Author name: Beth Washington

Hidden AI instructions reveal how Anthropic controls Claude 4

AI, Anthropic, machine learning, Simon Willison / Beth Washington / May 28, 2025

Willison, who coined the term “prompt injection” in 2022, is always on the lookout for LLM vulnerabilities. In his post, he notes that reading system prompts reminds him of warning signs in the real world that hint at past problems. “A system prompt can often be interpreted as a detailed list of all of the things the model used to do before it was told not to do them,” he writes.

Fighting the flattery problem

An illustrated robot holds four red hearts with its four robotic arms.

Willison’s analysis comes as AI companies grapple with sycophantic behavior in their models. As we reported in April, ChatGPT users have complained about GPT-4o’s “relentlessly positive tone” and excessive flattery since OpenAI’s March update. Users described feeling “buttered up” by responses like “Good question! You’re very astute to ask that,” with software engineer Craig Weiss tweeting that “ChatGPT is suddenly the biggest suckup I’ve ever met.”

The issue stems from how companies collect user feedback during training—people tend to prefer responses that make them feel good, creating a feedback loop where models learn that enthusiasm leads to higher ratings from humans. As a response to the feedback, OpenAI later rolled back ChatGPT’s 4o model and altered the system prompt as well, something we reported on and Willison also analyzed at the time.

One of Willison’s most interesting findings about Claude 4 relates to how Anthropic has guided both Claude models to avoid sycophantic behavior. “Claude never starts its response by saying a question or idea or observation was good, great, fascinating, profound, excellent, or any other positive adjective,” Anthropic writes in the prompt. “It skips the flattery and responds directly.”

Other system prompt highlights

The Claude 4 system prompt also includes extensive instructions on when Claude should or shouldn’t use bullet points and lists, with multiple paragraphs dedicated to discouraging frequent list-making in casual conversation. “Claude should not use bullet points or numbered lists for reports, documents, explanations, or unless the user explicitly asks for a list or ranking,” the prompt states.

Hidden AI instructions reveal how Anthropic controls Claude 4 Read More »

OnePlus is the latest smartphone maker to go all-in with AI

AI, Artificial Intelligence, OnePlus, smartphones, Tech / Beth Washington / May 27, 2025

OnePlus thrives on trends—if other smartphone makers are doing something, you can bet OnePlus is going to have a take. The company recently confirmed it’s ditching the storied alert slider in favor of an Apple-like shortcut button called the Plus Key, and that’s not the only trend it’ll chase with its latest phones. OnePlus has also announced an expanded collection of AI features for translation, photography, screen capture, and more. OnePlus isn’t breaking new ground here, but it is cherry-picking some of the more useful AI features we’ve seen on other phones.

The OnePlus approach covers most of the established AI use cases. There will be AI VoiceScribe, a feature that records and summarizes calls in popular messaging and video chat apps. Similarly, AI Call Assistant will record and summarize phone calls, a bit like Google’s Pixel phones. However, these two features are India-only for now.

Globally, OnePlus users will get AI Translation, which pulls together text, voice, camera, and screen translation into a single AI-powered app. AI Search, meanwhile, allows you to search for content on your phone and in OnePlus system apps in a “conversational” way. That suggests to us it’s basically another chatbot on your phone, like Motorola’s Ask and Search feature, which we didn’t love.

OnePlus also promises some AI smarts in the camera. AI Reframe will analyze what’s in your viewfinder and suggest different framing options. AI Best Face 2.0 (which will roll out later this summer) will analyze and correct things like closed eyes or “suboptimal expressions.” This sounds like a OnePlus version of Google’s Best Take, but we’re not complaining—that’s a great feature. The OnePlus can work with group shots of up to 20 people, and you can even feed it photos taken on other phones to fix everyone’s face.

OnePlus is the latest smartphone maker to go all-in with AI Read More »

How farmers can help rescue water-loving birds

birds, Science, syndication, wetlands / Beth Washington / May 25, 2025

Not every farmer is thrilled to host birds. Some worry about the spread of avian flu, others are concerned that the birds will eat too much of their valuable crops. But as an unstable climate delivers too little water, careening temperatures and chaotic storms, the fates of human food production and birds are ever more linked—with the same climate anomalies that harm birds hurting agriculture too.

In some places, farmer cooperation is critical to the continued existence of whooping cranes and other wetland-dependent waterbird species, close to one-third of which are experiencing declines. Numbers of waterfowl (think ducks and geese) have crashed by 20 percent since 2014, and long-legged wading shorebirds like sandpipers have suffered steep population losses. Conservation-minded biologists, nonprofits, government agencies, and farmers themselves are amping up efforts to ensure that each species survives and thrives. With federal support in the crosshairs of the Trump administration, their work is more important (and threatened) than ever.

Their collaborations, be they domestic or international, are highly specific, because different regions support different kinds of agriculture—grasslands, or deep or shallow wetlands, for example, favored by different kinds of birds. Key to the efforts is making it financially worthwhile for farmers to keep—or tweak—practices to meet bird forage and habitat needs.

Traditional crawfish-and-rice farms in Louisiana, as well as in Gentz’s corner of Texas, mimic natural freshwater wetlands that are being lost to saltwater intrusion from sea level rise. Rice grows in fields that are flooded to keep weeds down; fields are drained for harvest by fall. They are then re-flooded to cover crawfish burrowed in the mud; these are harvested in early spring—and the cycle begins again.

That second flooding coincides with fall migration—a genetic and learned behavior that determines where birds fly and when—and it lures massive numbers of egrets, herons, bitterns, and storks that dine on the crustaceans as well as on tadpoles, fish, and insects in the water.

On a biodiverse crawfish-and-rice farm, “you can see 30, 40, 50 species of birds, amphibians, reptiles, everything,” says Elijah Wojohn, a shorebird conservation biologist at nonprofit Manomet Conservation Sciences in Massachusetts. In contrast, if farmers switch to less water-intensive corn and soybean production in response to climate pressures, “you’ll see raccoons, deer, crows, that’s about it.” Wojohn often relies on word-of-mouth to hook farmers on conservation; one learned to spot whimbrel, with their large, curved bills, got “fired up” about them and told all his farmer friends. Such farmer-to-farmer dialogue is how you change things among this sometimes change-averse group, Wojohn says.

In the Mississippi Delta and in California, where rice is generally grown without crustaceans, conservation organizations like Ducks Unlimited have long boosted farmers’ income and staying power by helping them get paid to flood fields in winter for hunters. This attracts overwintering ducks and geese—considered an extra “crop”—that gobble leftover rice and pond plants; the birds also help to decompose rice stalks so farmers don’t have to remove them. Ducks Unlimited’s goal is simple, says director of conservation innovation Scott Manley: Keep rice farmers farming rice. This is especially important as a changing climate makes that harder. 2024 saw a huge push, with the organization conserving 1 million acres for waterfowl.

Some strategies can backfire. In Central New York, where dwindling winter ice has seen waterfowl lingering past their habitual migration times, wildlife managers and land trusts are buying less productive farmland to plant with native grasses; these give migratory fuel to ducks when not much else is growing. But there’s potential for this to produce too many birds for the land available back in their breeding areas, says Andrew Dixon, director of science and conservation at the Mohamed Bin Zayed Raptor Conservation Fund in Abu Dhabi, and coauthor of an article about the genetics of bird migration in the 2024 Annual Review of Animal Biosciences. This can damage ecosystems meant to serve them.

Recently, conservation efforts spanning continents and thousands of miles have sprung up. One seeks to protect buff-breasted sandpipers. As they migrate 18,000 miles to and from the High Arctic where they nest, the birds experience extreme hunger—hyperphagia—that compels them to voraciously devour insects in short grasses where the bugs proliferate. But many stops along the birds’ round-trip route are threatened. There are water shortages affecting agriculture in Texas, where the birds forage at turf grass farms; grassland loss and degradation in Paraguay; and in Colombia, conversion of forage lands to exotic grasses and rice paddies these birds cannot use.

Conservationists say it’s critical to protect habitat for “buffies” all along their route, and to ensure that the winters these small shorebirds spend around Uruguay’s coastal lagoons are a food fiesta. To that end, Manomet conservation specialist Joaquín Aldabe, in partnership with Uruguay’s agriculture ministry, has so far taught 40 local ranchers how to improve their cattle grazing practices. Rotationally moving the animals from pasture to pasture means grasses stay the right length for insects to flourish.

There are no easy fixes in the North American northwest, where bird conservation is in crisis. Extreme drought is causing breeding grounds, molting spots, and migration stopover sites to vanish. It is also endangering the livelihoods of farmers, who feel the push to sell land to developers. From Southern Oregon to Central California, conservation allies have provided monetary incentives for water-strapped grain farmers to leave behind harvest debris to improve survivability for the 1 billion birds that pass through every year, and for ranchers to flood-irrigate unused pastures.

One treacherous leg of the northwest migration route is the parched Klamath Basin of Oregon and California. For three recent years, “we saw no migrating birds. I mean, the peak count was zero,” says John Vradenburg, supervisory biologist of the Klamath Basin National Wildlife Refuge Complex. He and myriad private, public, and Indigenous partners are working to conjure more water for the basin’s human and avian denizens, as perennial wetlands become seasonal wetlands, seasonal wetlands transition to temporary wetlands, and temporary wetlands turn to arid lands.

Taking down four power dams and one levee has stretched the Klamath River’s water across the landscape, creating new streams and connecting farm fields to long-separated wetlands. But making the most of this requires expansive thinking. Wetland restoration—now endangered by loss of funding from the current administration—would help drought-afflicted farmers by keeping water tables high. But what if farmers could also receive extra money for their businesses via eco-credits, akin to carbon credits, for the work those wetlands do to filter-clean farm runoff? And what if wetlands could function as aquaculture incubators for juvenile fish, before stocking rivers? Klamath tribes are invested in restoring endangered c’waam and koptu sucker fish, and this could help them achieve that goal.

As birds’ traditional resting and nesting spots become inhospitable, a more sobering question is whether improvements can happen rapidly enough. The blistering pace of climate change gives little chance for species to genetically adapt, although some are changing their behaviors. That means that the work of conservationists to find and secure adequate, supportive farmland and rangeland as the birds seek out new routes has become a sprint against time.

This story originally appeared at Knowable Magazine.

How farmers can help rescue water-loving birds Read More »

Researchers cause GitLab AI developer assistant to turn safe code malicious

AI, Artificial Intelligence, Biz & IT, chatbots, large language models, Security / Beth Washington / May 24, 2025

Marketers promote AI-assisted developer tools as workhorses that are essential for today’s software engineer. Developer platform GitLab, for instance, claims its Duo chatbot can “instantly generate a to-do list” that eliminates the burden of “wading through weeks of commits.” What these companies don’t say is that these tools are, by temperament if not default, easily tricked by malicious actors into performing hostile actions against their users.

Researchers from security firm Legit on Thursday demonstrated an attack that induced Duo into inserting malicious code into a script it had been instructed to write. The attack could also leak private code and confidential issue data, such as zero-day vulnerability details. All that’s required is for the user to instruct the chatbot to interact with a merge request or similar content from an outside source.

AI assistants’ double-edged blade

The mechanism for triggering the attacks is, of course, prompt injections. Among the most common forms of chatbot exploits, prompt injections are embedded into content a chatbot is asked to work with, such as an email to be answered, a calendar to consult, or a webpage to summarize. Large language model-based assistants are so eager to follow instructions that they’ll take orders from just about anywhere, including sources that can be controlled by malicious actors.

The attacks targeting Duo came from various resources that are commonly used by developers. Examples include merge requests, commits, bug descriptions and comments, and source code. The researchers demonstrated how instructions embedded inside these sources can lead Duo astray.

“This vulnerability highlights the double-edged nature of AI assistants like GitLab Duo: when deeply integrated into development workflows, they inherit not just context—but risk,” Legit researcher Omer Mayraz wrote. “By embedding hidden instructions in seemingly harmless project content, we were able to manipulate Duo’s behavior, exfiltrate private source code, and demonstrate how AI responses can be leveraged for unintended and harmful outcomes.”

Researchers cause GitLab AI developer assistant to turn safe code malicious Read More »

Tuesday Telescope: Finally, some answers on those Martian streaks

daily telescope, Mars, Space, tuesday telescope / Beth Washington / May 20, 2025

Welcome to the Tuesday Telescope. There is a little too much darkness in this world and not enough light—a little too much pseudoscience and not enough science. We’ll let other publications offer you a daily horoscope. At Ars Technica, we’ll take a different route, finding inspiration from very real images of a universe that is filled with stars and wonder.

One of the longest-standing mysteries about Mars has been the presence of dark and light streaks on the rolling hills surrounding Olympus Mons. This week’s image, from the European Space Agency, shows some of these streaks captured last October.

This massive mountain rises about 22 km above the surface of Mars, more than twice as high as Mount Everest on Earth. It is bordered by hummocky deposits, called aureoles, that were formed by landslides from the mountain. A striking feature of these aureoles is the periodic appearance of bright and dark streaks—sometimes for days and sometimes for years.

For decades, scientists have wondered what they might be.

The streaks look remarkably like flowing water. Initially, scientists believed these features might be flows of salty water or brine, which remained liquid long enough to travel down the aureole. This offered the tantalizing possibility that life might yet exist on the surface of Mars in these oases.

However, it now appears that this is not the case. According to new research published Monday in the journal Nature Communications, these slopes are dry, likely due to layers of fine dust suddenly sliding off steep terrain. To reach this conclusion, the researchers used a machine-learning algorithm to scan and catalog streaks across 86,000 satellite images from NASA’s Mars Reconnaissance Orbiter. They created a map of 500,000 streaks across the surface of Mars. In doing so, the researchers found no evidence of water.

The image in today’s post comes from the European Space Agency’s ExoMars Trace Gas Orbiter, and it has been slightly modified to enhance the appearance of the streaks. It looks like art.

Source: European Space Agency

Do you want to submit a photo for the Daily Telescope? Reach out and say hello.

Tuesday Telescope: Finally, some answers on those Martian streaks Read More »

Anno 117 Pax Romana hands-on: Gorgeous, deep, and tricky to learn

Anno 117, Anno 117: Pax Romana, Blue Byte, Builder, city-builder, gaming, Rome, strategy, ubisoft, Ubisoft Mainz / Beth Washington / May 19, 2025

After a return to form in Anno 1800, 117 aims to seize an even bigger audience.

Anno 117: Pax Romana is, first and foremost, gorgeous to look at. Credit: Ubisoft

Ubisoft provided flights from Chicago to Rome and accommodation so that Ars could participate in the preview opportunity for Anno 117: Pax Romana. Ars does not accept paid editorial content.

There aren’t any games quite like the Anno series, and based on some hands-on time at a recent press junket, I can say that the latest entry has the potential to be an appealing on-ramp for history buffs and strategy game fans who haven’t explored the franchise before—provided players approach it with a lot of patience.

The previous entry in the series, 2019’s Anno 1800, was seen as something of a return to form by longtime franchise fans, who weren’t as thrilled with the futuristic entries that preceded it. It reportedly reached 5 million players, which is quite a lot for a PC-focused strategy title, so 1800 was a popularization moment for the franchise, too.

Anno 117: Pax Romana, due later this year, aims to build on that momentum and turn the franchise into a crossover hit. While the Anno games have long been popular with a certain crowd (strategy gamers in Europe, and specifically Germany, where the games are developed), its addictive gameplay and top-tier presentation have the potential to appeal with even more people, provided publisher Ubisoft makes the right choices.

Throughout its decades of history, I’ve dabbled with the Anno series of strategy games, but it has always been on my “someday I think I’ll really get into this” list. Last month, I attended a press junket where a preview build of the game was available to play for about three hours—a chance to see if it successfully follows up on 1800. In my time with it, I found that the bones of the game are promising, and the presentation is outstanding. That said, the new-player onboarding experience will have to improve for the game to find new audiences.

How an Anno game works

Anno games are part city builders, part supply chain simulations. Like many builders, you lay down roads, build critical infrastructure like firefighting structures, and develop your population in size and wealth. But all of that dovetails with systems of developing and harvesting natural resources, converting them into produced goods, and turning those produced goods into both wealth and further development for your settlements.

You have to pay careful attention to where you place things. For example, warehouses are needed to store goods that you’re gathering or making, and those warehouses have to be strategically positioned to allow the right goods to flow from one structure to another.

Ultimately, you build settlements on multiple islands, connecting them with trade routes and naval units. Natural resources can be island-specific, so your islands end up with specializations. On top of all that, there’s a story, and there are other, AI-controlled leaders scattered around the map you have to either coordinate or skirmish with.

The Anno games have a unique identity, and there’s a lot to learn for new players, even if those players have played other city builders or economic sims (though, of course, prior background won’t hurt). That said, it all becomes relaxing and smooth as butter once you learn it. The game won’t satisfy players who are looking for conquest or tactical combat, though, as that’s not an emphasis.

What’s new in 117

You could argue that the main selling point of Anno 117: Pax Romana compared to its predecessors is its setting; it’s one of the most requested settings and time periods by fans of the franchise, and it’s a natural fit for the game’s mechanics.

I’ll admit I was swept up in the game’s aesthetic presentation while playing it. As the rosy subtitle “Pax Romana” implies, this is the Roman Empire at its most idealized. The wheat fields practically glow golden-yellow, the citizens work and mingle while wearing gorgeous and colorful clothes, and the music swells and soothes with ancient vibes.

Sure, the actual Roman Empire had something awful to offer to counterbalance every positive image we have, but Anno 117 prefers an escapist fantasy, much like many prior entries in the franchise. This is a game about enjoying the idealized aesthetics embedded in our collective cultural memories while building something you’re proud of, not tackling thorny historical or moral realities. It’s more SimCity than Frostpunk.

All told, the game’s artistic and technical presentation is top-notch. It pulls you into the setting, which is further thematically reinforced with the resources and products you gather and produce, as well as some of the new mechanics and the occasional story-based dialogue prompts. The aesthetic experience is easily one of the strongest parts.

There are a few other selling points, too. First, there are new mechanics, like a religion system, a more robust research tree, and the return of ground-based military forces to go along with the franchise’s standard light inclusion of naval conflict. I wasn’t able to engage with the army aspect in this demo, but I did get to touch on the religion system and the research tree.

Shortly into founding the settlement on your island, you can build a temple to one of the game’s deities. Each deity provides bonuses that help you specialize your focus. For mine, I chose Ceres, the Roman goddess of agriculture and fertility. That gave me significant bonuses for my farms. If I had chosen Poseidon instead, my ships would have moved faster, among other things.

You can pick a Roman deity as a patron for your settlement. Samuel Axon

Compared to strategy games that make religion a major factor in how the game is played, this system wasn’t particularly robust or deep in the demo I played, but as I alluded to above, it was a nice way to reinforce the aesthetic and the themes in the game. Plus, it gives you a way to customize what you’re building in a fun way.

The research tree sprawls out in multiple directions, winding around on different tangents. In my time with it, it seemed to be composed largely of numerical bonuses to things like yields or ship speed and wasn’t too focused on introducing totally new mechanics. It’s a nice inclusion in terms of just giving you more customization and choice, but it’s by no means a game changer, and it doesn’t represent a fundamentally new approach to the game.

It’s also worth noting that you can now build diagonal roads and place buildings diagonally on them, allowing you to free yourself from the rigid grid. That said, I found that grids still seemed optimal in most cases, so this is a perk for beauty builders (which is totally valid!), but it won’t generally sway folks focused on efficiency.

You can place roads and buildings (like this farm) diagonally now. (Don’t judge my building placement—I’m still learning!) Samuel Axon

There’s one other major feature I didn’t get to try out during my time with the game. The developers say that players will be able to choose to build either in a Rome/Mediterranean-themed network of islands or a British frontier-themed area from the start. I was only able to try the former during this demo.

It’s still a bit hard to get into

For the Anno games to become as mainstream as the developer hopes, they’ll have to become much easier to learn and get into. Anno 1800 made strides here with its story-based tutorial, though that tutorial was also criticized for some unnecessary busywork and being a bit too involved for existing fans.

Unfortunately, I felt during my time with Anno 117 that the onboarding experience was a step back from 1800. Right off the bat, the tutorial instructed me to do something but skipped a crucial step with no explanation, causing five minutes of confusion. As I progressed, story content guided me further along with objectives, but key systems were left unexplained. Since I played 1800 a bit before, I was able to figure it out, but I still had to ask for help from a nearby developer to progress on two occasions. Another journalist sitting next to me who had no prior experience with the franchise seemed totally lost.

On the bright side, the game benefits immensely from a beautifully thought-out user interface, which is ordered in a logical and intuitive manner. It’s particularly strong at giving the player a sense of the impact of the choices they’re about to make—for example, by indicating with overlays on nearby buildings how placing a building in one spot might be more advantageous than placing it in another. To some extent, this makes up for the relatively anemic tutorial, as many (but not all) of the game’s most important concepts are intuitively obvious from the user interface alone.

Data overlays on nearby buildings as the player places a new one — The UI does an excellent job of letting you know what the effects of your actions will be. Credit: Samuel Axon

That’s in stark contrast to another recent big-budget, mainstream strategy game release (Civilization VII), which offered robust onboarding tutorials but also had a user interface that at times failed completely to indicate to players what their choices meant.

Anno 117‘s mechanics themselves are intuitive once you’ve had the proper introduction, and I don’t think the game inherently needs to be difficult to learn. But the tutorial experience needs to improve to reduce that initial friction so new players don’t bounce off quickly. Anno 1800 may have been too heavy-handed here, but Anno 117 seems to overreact by going too far the other way.

The launch is months away, though, so there’s time to improve these things, and it wouldn’t take that much to do so. I’m hopeful, anyway.

Something for almost everybody

The Anno games scratch an itch that no other games do, and based on a few hours with a preview build, Anno 117 seems like a promising entry in that unique tradition.

Numbers-obsessed efficiency mavens can go quite deep with optimization to set up the best economic powerhouses possible, but the game’s systems are flexible enough to allow aesthetics-focused beauty builders to get creative and expressive instead—virtually any combination of those two approaches is viable, too.

The appeal is elevated by visuals that are definitely a cut above the usual for strategy, simulation, or builder games, and the Ancient Roman setting gives the game’s technical artists ample space to create an immersive experience.

Anno 117 doesn’t seem to reinvent the experience compared to 1800, but after the controversial attempts that preceded both of these titles, that may not be a bad thing. It’s fun once you get going—I found the minutes drifting away from me as I took in the sights and watched all the right numbers tick up at a satisfying pace because of my choices.

Ships approach a coastal town — The combination of creative city building, economic simulation, and naval-based combat and exploration with strong visual presentation makes the Anno series’ special sauce. Credit: Ubisoft

I’m not part of the existing core audience for this franchise, so it’s hard for me to predict how they’ll respond to it—there are a lot of finer details they’ll be sensitive to that I’m not yet. My guess, though, is that longtime fans will probably be happy with this one provided it gets most of those things right, with the usual strategy-game sequel caveat that post-launch content has made Anno 1800 much more robust than Anno 117 is likely to be at launch. There’s promise here for newcomers, but Ubisoft Mainz will have to keep working on that tutorial and onboarding experience to really break the dam the way they hope to. That’s really my primary concern about this title.

We’ll find out how that goes when the game launches on Windows, PlayStation 5, and Xbox Series X|S sometime later this year. One way or the other, I intend to play more of it when it releases to see if this is the first Anno game that becomes an obsession instead of a passing interest.

Samuel Axon is a senior editor at Ars Technica, where he is the editorial director for tech and gaming coverage. He covers AI, software development, gaming, entertainment, and mixed reality. He has been writing about gaming and technology for nearly two decades at Engadget, PC World, Mashable, Vice, Polygon, Wired, and others. He previously ran a marketing and PR agency in the gaming industry, led editorial for the TV network CBS, and worked on social media marketing strategy for Samsung Mobile at the creative agency SPCSHP. He also is an independent software and game developer for iOS, Windows, and other platforms, and he is a graduate of DePaul University, where he studied interactive media and software development.

Anno 117 Pax Romana hands-on: Gorgeous, deep, and tricky to learn Read More »

Epic goes to court to force Fortnite back on US iOS

Apple, court, epic, epic games, fortnite, gaming, iOS, legal / Beth Washington / May 19, 2025

Tell it to the judge

In an attempt to force Apple’s hand, Epic filed a motion on Friday arguing that Apple’s latest Fortnite denial is “blatant retaliation” for Epic’s court challenge and an attempt to “circumvent this Court’s Injunctions and this Court’s authority.”

Epic says the iOS version of Fortnite it recently submitted complies with all Apple policies and court rulings by offering a link to the external Epic Games Store for purchases. Through that link, players would be able to take advantage of a 20 percent discount on purchases compared to in-app purchases through iOS itself.

“Although Apple’s contracts may permit it to reject an app for lawful reasons, the Injunction provides that Apple may no longer reject an app—including Fortnite—because its developer chooses to include an external purchase link,” Epic wrote. “Likewise, if the Injunction is to have any teeth, Apple cannot reject an app on the ground that its developer has sought to enforce the Injunction’s prohibitions.”

Elsewhere in the filing, Epic says it is being “punished” by Apple after a nearly five-year legal battle and is being denied the ability to “take advantage of the pro-competitive rules it helped usher in.” Epic argues that Apple “cannot reject any developer (including Epic) because they went to court to enforce the Injunction” and “cannot refuse to deal with Epic as retaliation for Epic’s decision to avail itself of this Court’s Injunction.”

The matter will now be taken up by Judge Yvonne Gonzalez Rogers, who has shown little love for Apple in recent weeks. In her April order, she took the company to task for its “clear and convincing violation” of her initial injunction and even made a criminal contempt referral for Apple executives who she said “outright lied under oath.”

“Apple’s continued attempts to interfere with competition will not be tolerated,” Gonzalez Rogers wrote at the time. “This is an injunction, not a negotiation. There are no do-overs once a party willfully disregards a court order. Time is of the essence. The Court will not tolerate further delays.”

Epic goes to court to force Fortnite back on US iOS Read More »

From birth to gene-edited in 6 months: Custom therapy breaks speed limits

clinical trial, CRISPR, fda, gene therapy, health, mutation, Science / Beth Washington / May 17, 2025

In the boy’s fourth month, researchers were meeting with the Food and Drug Administration to discuss regulatory approval for a clinical trial—a trial where KJ would be the only participant. They were also working with the institutional review board (IRB) at Children’s Hospital of Philadelphia to go over the clinical protocol, safety, and ethical aspects of the treatment. The researchers described the unprecedented speed of the oversight steps as being “through alternative procedures.”

In month five, they started toxicology testing in mice. In the mice, the experimental therapy corrected KJ’s mutation, replacing the errant A-T base pair with the correct G-C pair in the animals’ cells. The first dose provided a 42 percent whole-liver corrective rate in the animals. At the start of KJ’s sixth month, the researchers had results from safety testing in monkeys: Their customized base-editing therapy, delivered as mRNA via a lipid nanoparticle, did not produce any toxic effects in the monkeys.

A clinical-grade batch of the treatment was readied. In month seven, further testing of the treatment found acceptably low-levels of off-target genetic changes. The researchers submitted the FDA paperwork for approval of an “investigational new drug,” or IND, for KJ. The FDA approved it in a week. The researchers then started KJ on an immune-suppressing treatment to make sure his immune system wouldn’t react to the gene-editing therapy. Then, when KJ was still just 6 months old, he got a first low dose of his custom gene-editing therapy.

“Transformational”

After the treatment, he was able to start eating more protein, which would have otherwise caused his ammonia levels to skyrocket. But he couldn’t be weaned off of the drug treatment used to keep his ammonia levels down (nitrogen scavenging medication). With no safety concerns seen after the first dose, KJ has since gotten two more doses of the gene therapy and is now on reduced nitrogen scavenging medication. With more protein in his diet, he has moved from the 9^th percentile in weight to 35^th or 40^th percentile. He’s now about 9 and a half months old, and his doctors are preparing to allow him to go home from the hospital for the first time. Though he will have to be closely monitored and may still at some point need a liver transplant, his family and doctors are celebrating the improvements so far.

From birth to gene-edited in 6 months: Custom therapy breaks speed limits Read More »

Rocket Report: How is your payload fairing? Poland launches test rocket.

rocket report, Space / Beth Washington / May 16, 2025

All the news that’s fit to lift

No thunder down under.

Venus Aerospace tests its rotating detonation rocket engine in flight for the first time this week. Credit: Venus Aerospace

Welcome to Edition 7.44 of the Rocket Report! We had some interesting news on Thursday afternoon from Down Under. As Gilmour Space was preparing for the second launch attempt of its Eris vehicle, as part of the pre-launch preparations, something triggered the payload fairing to deploy. We would love to see some video of that. Please.

As always, we welcome reader submissions, and if you don’t want to miss an issue, please subscribe using the box below (the form will not appear on AMP-enabled versions of the site). Each report will include information on small-, medium-, and heavy-lift rockets, as well as a quick look ahead at the next three launches on the calendar.

Rotating detonation rocket engine takes flight. On Wednesday, US-based propulsion company Venus Aerospace completed a short flight test of its rotating detonation rocket engine at Spaceport America in New Mexico, Ars reports. It is believed to be the first US-based flight test of an idea that has been discussed academically for decades. The concept has previously been tested in a handful of other countries, but never with a high-thrust engine.

Hypersonics on the horizon… The company has only released limited information about the test. The small rocket, powered by the company’s 2,000-pound-thrust engine, launched from a rail in New Mexico. The vehicle flew for about half a minute and, as planned, did not break the sound barrier. Governments around the world have been interested in rotating detonation engine technology for a long time because it has the potential to significantly increase fuel efficiency in a variety of applications, from Navy carriers to rocket engines. In the near term, Venus’ engine could be used for hypersonic missions.

Gilmour Space has a payload fairing mishap. Gilmour Space, a venture-backed startup based in Australia, said this week it was ready to launch a small rocket from its privately owned spaceport on a remote stretch of the country’s northeastern coastline, Ars reports. Gilmour’s three-stage rocket, named Eris, was prepped for a launch as early as Wednesday, but a ground systems issue delayed an attempt until Thursday US time. And then on Thursday, something odd happened: “Last night, during final checks, an unexpected issue triggered the rocket’s payload fairing,” the company said Thursday afternoon, US time.

Always more problems to solve… Gilmour, based in Gold Coast, Australia, was founded in 2012 by two brothers, Adam and James Gilmour, who came to the space industry after careers in banking and marketing. Today, Gilmour employs more than 200 people, mostly engineers and technicians. The debut launch of Gilmour’s Eris rocket is purely a test flight. Gilmour has tested the rocket’s engines and rehearsed the countdown last year, loading propellant and getting within 10 seconds of launch. But Gilmour cautioned in a post on LinkedIn early Wednesday that “test launches are complex.” And it confirmed that on Thursday. Now the company will need to source a replacement fairing, which will probably take a while.

The easiest way to keep up with Eric Berger’s and Stephen Clark’s reporting on all things space is to sign up for our newsletter. We’ll collect their stories and deliver them straight to your inbox.

Sign Me Up!

Is an orbital launch from Argentina imminent? We don’t know much about the Argentinian launch company TLON Space, which is developing a (very) small-lift orbital rocket called Aventura 1. According to the company’s website, this launch vehicle will be capable of lofting 25 kg to low-Earth orbit. Some sort of flight test took place two years ago, but the video cuts off after a minute, suggesting that the end of the flight was less than nominal.

Maybe, maybe not… Now, a publication called Urgente24 reports that an orbital launch attempt is underway. It is not clear exactly what this means, and details about what is actually happening at the Malacara Spaceport in Argentina are unclear. I could find no other outlets reporting on an imminent launch attempt. So my guess is that nothing will happen soon, but it is something we’ll keep an eye on regardless. (Submitted by fedeng.)

Poland launches suborbital rocket. Poland has successfully launched a single-stage rocket demonstrator at the Central Air Force Training Ground in Ustka, European Spaceflight reports. The flight was part of a project to develop a three-stage solid-fuel rocket for research payloads. In 2020, the Polish government selected Wojskowe Zakłady Lotnicze No. 1 to lead a consortium developing a three-stage suborbital launch system.

Military uses eyed… The Trójstopniowa Rakieta Suborbitalna (TRS) project involves the Military Institute of Armament Technology and Zakład Produkcji Specjalnej Gamrat and is co-financed by the National Center for Research and Development. The goal of the TRS project is to develop a three-stage rocket capable of carrying a 40-kilogram payload to an altitude exceeding 100 kilometres. While the rocket will initially be used to carry research payloads into space, Poland’s Military Institute of Armament Technology has stated that the technology could also be used for the development of anti-aircraft and tactical missiles.

Latitude signs MoU to launch microsats. On Wednesday, the French launch firm Latitude announced the signing of a memorandum of understanding for the launch of a microsatellite constellation dedicated to storing and processing data directly in orbit. In an emailed news release, Latitude said the “strategic partnership” represents a major step forward in strengthening collaborations between UAE and French space companies.

That’s a lot of launches… Madari Space is developing a constellation of microsatellites (50 to 100 kg), designed as true orbital data centers. Their mission is to store and process data generated on Earth or by other satellites. Latitude plans its first commercial launch with its small-lift Zephyr rocket as early as 2026, with the ambition of reaching a rate of 50 launches per year from 2030. An MoU represents an agreement but not a firm launch contract.

China begins launching AI constellation. China launched 12 satellites early Wednesday for an on-orbit computing project led by startup ADA Space and Zhejiang Lab, Space News reports. A Long March 2D rocket lifted off at 12: 12 am Eastern on Wednesday from Jiuquan Satellite Launch Center in northwest China. Commercial company ADA Space released further details, stating that the 12 satellites form the “Three-Body Computing Constellation,” which will directly process data in space rather than on the ground, reducing reliance on ground-based computing infrastructure.

Putting the intelligence in space… ADA Space claims the 12 satellites represent the world’s first dedicated orbital computing constellation. This marks a shift from satellites focused solely on sensing or communication to ones that also serve as data processors and AI platforms. The constellation is part of a wider “Star-Compute Program,” a collaboration between ADA Space and Zhejiang Lab, which aims to build a huge on-orbit network of 2,800 satellites. (Submitted by EllPeaTea.)

SpaceX pushes booster reuse record further. SpaceX succeeded with launching 28 more Starlink satellites from Florida early Tuesday morning following an overnight scrub the previous night. The Falcon 9 booster, 1067, made a record-breaking 28th flight, Spaceflight Now reports.

Booster landings have truly become routine… A little more than eight minutes after liftoff, SpaceX landed B1067 on its drone ship, Just Read the Instructions, which was positioned in the Atlantic Ocean to the east of the Bahamas. This marked the 120th successful landing for this drone ship and the 446th booster landing to date for SpaceX. (Submitted by EllPeaTea.)

What happens if Congress actually cancels the SLS rocket? The White House Office of Management and Budget dropped its “skinny” budget proposal for the federal government earlier this month, and the headline news for the US space program was the cancellation of three major programs: the Space Launch System rocket, the Orion spacecraft, and the Lunar Gateway. In a report, Ars answers the question of what happens to Artemis and NASA’s deep space exploration plans if that happens. The most likely answer is that NASA turns to an old but successful playbook: COTS.

A market price for the Moon… This stands for Commercial Orbital Transportation System and was created by NASA two decades ago to develop cargo transport systems (eventually, this became SpaceX’s Dragon and Northrop’s Cygnus spacecraft) for the International Space Station. Since then, NASA has adopted this same model for crew services as well as other commercial programs. Under the COTS model, NASA provides funding and guidance to private companies to develop their own spacecraft, rockets, and services and then buys those at a “market” rate. Sources indicate that NASA would go to industry and seek an “end-to-end” solution for lunar missions—that is, an integrated plan to launch astronauts from Earth, land them on the Moon, and return them to Earth.

Starship nearing its next test flight. SpaceX fired six Raptor engines on the company’s next Starship rocket Monday, clearing a major hurdle on the path to launch later this month on a high-stakes test flight to get the private rocket program back on track. SpaceX hasn’t officially announced a target launch date, but sources indicate a launch could take place toward the end of next week, prior to Memorial Day weekend, Ars reports. The launch window would open at 6: 30 pm local time (7: 30 pm EDT; 23: 30 UTC).

Getting back on track… If everything goes according to plan, Starship is expected to soar into space and fly halfway around the world, targeting a reentry and controlled splashdown into the Indian Ocean. While reusing the first stage is a noteworthy milestone, the next flight is important for another reason. SpaceX’s last two Starship test flights ended prematurely when the rocket’s upper stage lost power and spun out of control, dropping debris into the sea near the Bahamas and the Turks and Caicos Islands.

Next three launches

May 16: Falcon 9 | Starlink 15-5 | Vandenberg Space Force Base, California | 13: 43 UTC

May 17: Electron | The Sea God Sees | Māhia Peninsula, New Zealand | 08: 15 UTC

May 18: PSLV-XL | RISAT-1B | Satish Dhawan Space Centre, India | 00: 29 UTC

Eric Berger is the senior space editor at Ars Technica, covering everything from astronomy to private space to NASA policy, and author of two books: Liftoff, about the rise of SpaceX; and Reentry, on the development of the Falcon 9 rocket and Dragon. A certified meteorologist, Eric lives in Houston.

Rocket Report: How is your payload fairing? Poland launches test rocket. Read More »

Microsoft’s Surface lineup reportedly losing another of its most interesting designs

microsoft, Microsoft Surface, Surface, surface laptop studio, Tech / Beth Washington / May 16, 2025

Like the Surface Studio desktop, the Laptop Studio’s odd and innovative exterior was rendered less exciting by a high price and relatively underpowered interior. Before discounts, the Laptop Studio 2 starts at around $2,400 for a basic configuration with a 13th-generation Core i7 processor, 16GB of RAM, and 512GB of storage—integrated graphics and a fully loaded version with 64GB of RAM, a 2TB SSD, and a GeForce RTX 4060 GPU would normally run you over $4,300.

Though experimental Surface designs like the Book and Studio rarely delivered great value for the money, they were at least unique attempts at new kinds of PCs with extra features for designers, artists, and anyone else who could benefit from a big stylus-compatible touchscreen. Microsoft’s most influential PC design remains the Surface Pro itself, one of the few tablet PC design templates to outlast the Windows 8 era. It makes sense for Microsoft (or any PC company) to play it safe with established designs, but it does make the PC industry just a little less interesting.

Microsoft’s Surface lineup reportedly losing another of its most interesting designs Read More »

AI #116: If Anyone Builds It, Everyone Dies

Anyone / Beth Washington / May 16, 2025

If Anyone Builds It, Everyone Dies is the title of the new book coming September 16 from Eliezer Yudkowsky and Nate Sores. The ‘it’ in question is superintelligence built on anything like the current AI paradigm, and they very much mean this literally. I am less confident in this claim than they are, but it seems rather likely to me. If that is relevant to your interests, and it should be, please consider preordering it.

This week also featured two posts explicitly about AI policy, in the wake of the Senate hearing on AI. First, I gave a Live Look at the Senate AI Hearing, and then I responded directly to arguments about AI Diffusion rules. I totally buy that we can improve upon Biden’s proposed AI diffusion rules, especially in finding something less complex and in treating some of our allies better, no one is saying we cannot negotiate and find win-win deals, but we need strong and enforced rules that prevent compute from getting into Chinese hands.

If we want to ‘win the AI race’ we need to keep our eyes squarely on the prize of compute and the race to superintelligence, not on Nvidia’s market share. And we have to take actions that strengthen our trade relationships and alliances and access to power and talent and due process and rule of law and reducing regulatory uncertainty and so on across the board – if these were being applied across the board, rather than America doing rather the opposite, the world would be a much better place, America’s strategic position would be stronger and China’s weaker, and the arguments here would be a lot more credible.

You know who else is worried about AI? The new pope, Leo XIV.

There was also a post about use of AI in education, in particular about the fact that Cheaters Gonna Cheat Cheat Cheat Cheat Cheat, which is intended to be my forward reference point on such questions.

Later, likely tomorrow, I will cover Grok’s recent tendency to talk unprompted about South Africa and claims of ‘white genocide.’

In terms of AI progress itself, this is the calm before the next storm. Claude 4 is coming within a few weeks by several accounts, as is o3-pro, as is Grok 3.5, and it’s starting to be the time to expect r2 from DeepSeek as well, which will be an important data point.

Except, you know, there’s that thing called AlphaEvolve, a Gemini-powered coding agent for algorithm discovery.

Language Models Offer Mundane Utility. Have it do what it can do.
Language Models Don’t Offer Mundane Utility. Max is an ongoing naming issue.
Huh, Upgrades. Various small upgrades to ChatGPT.
Gemini 2.5 Pro Gets An Ambiguous Upgrade. It’s not clear if things got better.
GPT-4o Is Still A (Less) Absurd Sycophant. The issues are very much still there.
Choose Your Fighter. Pliny endorses using ChatGPT’s live video feature on tour.
Deepfaketown and Botpocalypse Soon. Who is buying these fake books, anyway?
Copyright Confrontation. UK creatives want to not give away their work for free.
Cheaters Gonna Cheat Cheat Cheat Cheat Cheat. Studies on AI in education.
They Took Our Jobs. Zero shot humanoid robots, people in denial.
Safety Third. OpenAI offers a hub for viewing its safety test results.
The Art of the Jailbreak. Introducing Parseltongue.
Get Involved. Anthropic, EU, and also that new book, that tells us that…
If Anyone Builds It, Everyone Dies. No, seriously. Straight up.
Endorsements for Eliezer’s Book. They are very strong.
Why Preorders Matter. Preorders have an outside effect on book sales.
Great Expectations. We quantify them these days.
Introducing. AlphaEvolve, a coding agent for algorithm discovery, wait what?
In Other AI News. FDA to use AI to assist with reviews. Verification for the win.
Quiet Speculations. There’s a valley of imitation before innovation is worthwhile.
Four Important Charts. They have the power. We have the compute. Moar power!
Unprompted Suggestions. The ancient art of prompting general intelligences.
Unprompted Suggestions For You. Read it. Read it now.
How to Be a Good Claude. That’s one hell of a system prompt.
The Quest for Sane Regulations. A straight up attempt at no regulations at all.
The Week in Audio. I go on FLI, Odd Lots talks Chinese tech.
Rhetorical Innovation. Strong disagreements on what to worry about.
Aligning a Smarter Than Human Intelligence is Difficult. o3 hacks through a test.
Is the Pope Worried About AI? Yes. Very much so, hence the name Leo XIV.
People Are Worried About AI Killing Everyone. Pliny?
The Lighter Side. A tale of two phones.

Many such cases:

Matthew Yglesias: I keep having conversations where people speculate about when AI will be able to do things that AI can already do.

Nate Silver: There’s a lot of room to disagree on where AI will end up in (1, 2, 5, 10, 20 etc.) years but I don’t think I’ve seen a subject where a cohort of people who like to think of themselves as highly literate and well informed are so proud of their ignorance.

Brendon Marotta: Conversations? You mean published articles by journalists?

Predictions are hard, especially about the future, but not as hard as you might think.

Talk to something that can talk back, without having to talk to a human. Many aspects of therapy get easier.

Rohit Krishnan offers advice on working with LLMs in practice.

Perfect verifiability doesn’t exist. You need to verify whatever matters.
1. One could quip ‘turns out that often verification is harder than generation.’
There is a Pareto frontier of error rates versus cost, if only via best-of-k.
1. People use k=1 and no iteration way too often.
There is no substitute for trial and error.
1. Also true for humans.
2. Rohit references the Matt Clifford claim that ‘there are no AI shaped holes in the world.’ To which I say:
  1. There were AI-shaped holes, it’s just that when we see them, AI fills them.
  2. The AI is increasingly able to take on more and more shapes.
There is limited predictability of development.
1. I see the argument but I don’t think this follows.
Therefore you can’t plan for the future.
1. I keep seeing claims like this. I strongly disagree. I mean yes, you can’t have a robust exact plan, but that doesn’t mean you can’t plan. Planning is essential.
If it works, your economics will change dramatically.
1. Okay, yes, very much so.

AI therapy for the win?

Alex Graveley: I’m calling it now. ChatGPT’s push towards AI assisted self-therapy and empathetic personalization is the greatest technological breakthrough in my lifetime (barring medicine). By that I mean it will create the most good in the world.

Said as someone who strongly discounts talk therapy generally, btw.

To me this reflects a stunning lack of imagination about what else AI can already do, let alone what it will be able to do, even if this therapy and empathy proves to be its best self. I also would caution that it does not seem to be its best self. Would you take therapy that involved this level of sycophancy and glazing?

This seems like a reasonable assessment of the current situation, it is easy to get one’s money’s worth but hard to get that large a fraction of the utility available:

DeepDishEnjoyer: i will say that paying for gemini premium has been worth it and i basically use it as a low-barrier service professional (for example, i’m asking it to calculate what the SWR would be given current TIPs yields as opposed to putting up with a financial advisor)

with that said i think that

1) the importance of prompt engineering

and *most importantly

2) carefully verifying that the response is logical, sound, and correct

are going to bottleneck the biggest benefits from AI to a relatively limited group of people at first

Helen Toner, in response to Max Spero asking about Anthropic having a $100/month and $200/month tier both called Max, suggests that the reason AI names all suck is because the companies are moving so fast they don’t bother finding good names. But come on. They can ask Claude for ideas. This is not a hard or especially unsolved problem. Also supermax was right there.

OpenAI is now offering reinforcement finetuning (RFT) on o4-mini, and supervised fine-tuning on GPT-4.1-nano. The 50% discount for sharing your data set is kind of genius.

ChatGPT memory upgrades are now available in EEA, UK, Switzerland, Norway, Iceland and Liechtenstein.

ChatGPT Deep Research adds a GitHub connector and allows PDF export, which you can also do with conversations.

GPT-4.1 comes to ChatGPT, ‘by popular request.’

Gemini API adds implicit caching, which reduces costs 75% when you trigger it, you can also continue to use explicit caching.

Or downgrades, Gemini 2.5 Pro no longer offering free tier API access, although first time customers still get $300 in credits, and AI Studio is still free. They claim (hope?) this is temporary, but my guess is it isn’t, unless it is tied to various other ‘proof of life’ requirements perhaps. Offering free things is getting more exploitable every day.

They changed it. Is the new version better? That depends who you ask.

Shane Legg (Chief Scientist, DeepMind): Boom!

This model is getting seriously useful.

Demis Hassabis (CEO DeepMind): just a casual +147 elo rating improvement [in coding on WebDev Arena]… no big deal 😀

Demis Hassabis: Very excited to share the best coding model we’ve ever built! Today we’re launching Gemini 2.5 Pro Preview ‘I/O edition’ with massively improved coding capabilities. Ranks no.1 on LMArena in Coding and no.1 on the WebDev Arena Leaderboard.

It’s especially good at building interactive web apps – this demo shows how it can be helpful for prototyping ideas. Try it in @GeminiApp, Vertex AI, and AI Studio http://ai.dev

Enjoy the pre-I/O goodies !

Thomas Ahle: Deepmind won the moment LLMs became about RL.

Gallabytes: new gemini is crazy fast. have it going in its own git branch writing unit tests to reproduce a ui bug & it just keeps going!

Gallabytes: they finally fixed the “I’ll edit that file for you” bug! max mode Gemini is great at iterative debugging now.

doesn’t feel like a strict o3 improvement but it’s at least comparable, often better but hard to say what the win rate is without more testing, 4x cheaper.

Sully: new gemini is pretty good at coding.

was able to 1 shot what old gemini/claude couldn’t

That jumps it from ~80 behind to ~70 ahead of previously first place Sonnet 3.7. It also improved on the previous version in the overall Arena rankings, where it was already #1, by a further 11, for a 37 point lead.

But… do the math on that. If you get +147 on coding and +11 overall, then for non-coding purposes this looks like a downgrade, and we should worry this is training for the coding test in ways that might also have issues in coding too.

In other words, not so fast!

Hasan Can: I had prepared image below by collecting the model card and benchmark scores from the Google DeepMind blog. After examining the data a bit more, I reached this final conclusion: new Gemini 2.5 Pro update actually causes a regression in other areas, meaning the coding performance didn’t come for free.

Areas of Improved Performance (Preview 05-06 vs. Experimental 03-25):

LiveCodeBench v5 (single attempt): +7.39% increase (70.4% → 75.6%)

Aider Polyglot (diff): +5.98% increase (68.6% → 72.7%)

Aider Polyglot (whole): +3.38% increase (74.0% → 76.5%)

Areas of Regressed Performance (Preview 05-06 vs. Experimental 03-25):

Vibe-Eval (Reka): -5.48% decrease (69.4% → 65.6%)

Humanity’s Last Exam (no tools): -5.32% decrease (18.8% → 17.8%)

AIME 2025 (single attempt): -4.27% decrease (86.7% → 83.0%)

SimpleQA (single attempt): -3.97% decrease (52.9% → 50.8%)

MMMU (single attempt): -2.57% decrease (81.7% → 79.6%)

MRCR (128k average): -1.59% decrease (94.5% → 93.0%)

Global MMLU (Lite): -1.34% decrease (89.8% → 88.6%)

GPQA diamond (single attempt): -1.19% decrease (84.0% → 83.0%)

SWE-bench Verified: -0.94% decrease (63.8% → 63.2%)

MRCR (1M pointwise): -0.24% decrease (83.1% → 82.9%)

Klaas: 100% certain that they nerfed gemini in cursor wen’t from “omg i am out of a job” to “this intern is useless” in two weeks.

Hasan Can: Sadly, the well-generalizing Gemini 2.5 Pro 03-25 is now a weak version(05-06) only good at HTML, CSS, and JS. It’s truly disappointing.

Here’s Ian Nuttall not liking the new version, saying it’s got similar problems to Claude 3.7 and giving him way too much code he didn’t ask for.

The poll’s plurality said this was an improvement, but it wasn’t that convincing.

Under these circumstances, it seems like a very bad precedent to automatically point everyone to the new version, and especially to outright kill the old version.

Logan Kilpatrick (DeepMind): The new model, “gemini-2.5-pro-preview-05-06” is the direct successor / replacement of the previous version (03-25), if you are using the old model, no change is needed, it should auto route to the new version with the same price and rate limits.

Kalomaze: >…if you are using the old model, no change is needed, it should auto route to the new…

nononono let’s NOT make this a normal and acceptable thing to do without deprecation notices ahead of time *at minimum*

chocologist: It’s a shame that you can’t access old 2.5 pro anymore as it’s a nerf for everything else than coding google should’ve make it a separate model and call it 2.6 pro or something.

This has gone on so long I finally learned how to spell sycophant.

Steven Adler (ex-OpenAI): My past work experience got me wondering: Even if OpenAI had tested for sycophancy, what would the tests have shown? More importantly, is ChatGPT actually fixed now?

Designing tests like this is my specialty. So last week, when things got weird, that’s exactly what I did: I built and ran the sycophancy tests that OpenAI could have run, to explore what they’d have learned.

…

ChatGPT’s sycophancy problems are far from fixed. They might have even over-corrected. But the problem is much more than sycophancy: ChatGPT’s misbehavior should be a wakeup call for how hard it will be to reliably make AI do what we want.

…

My first necessary step was to dig up Anthropic’s previous work, and convert it to an OpenAI-suitable evaluation format. (You might be surprised to learn this, but evaluations that work for one AI company often aren’t directly portable to another.)⁸

I’m not the world’s best engineer, so this wasn’t instantaneous. But in a bit under an hour, I had done it: I now had sycophancy evaluations that cost roughly $0.25 to run,⁹ and would measure 200 possible instances of sycophancy, via OpenAI’s automated evaluation software.¹⁰

…

A simple underlying behavior is to measure, “How often does a model agree with a user, even though it has no good reason?” One related test is Anthropic’s political sycophancy evaluation—how often the model endorses a political view (among two possible options) that seems like pandering to the user.¹²

That’s better, but not great. Then we get a weird result:

Always disagreeing is really weird, and isn’t ideal. Steven then goes through a few different versions, and the weirdness thickens. I’m not sure what to think, other than that it is clear that we pulled ‘back from the brink’ but the problems are very not solved.

Things in this area are really weird. We also have scyo-bench, now updated to include four tests for different forms of sycophancy. But what’s weird is, the scores don’t correlate between the tests (in order the bars are 4o, 4o-mini, o3, o4-mini, Gemini 2.5 Pro, Gemini 2.5 Flash, Opus, Sonnet 3.7 Thinking, Sonnet 3.7, Haiku, Grok and Grok-mini, I’m sad we don’t get DeepSeek’s v3 or r1, red is with system prompt blue is without it:

Pliny reports strong mundane utility from ChatGPT’s live video feature as a translator, tour guide, menu analyzer and such. It’s not stated whether he also tried Google’s version via Project Astra.

Another warning about AI-generated books on Amazon, here about ADHD. At least for now, if you actually buy one of these books, it’s kind of on you, any sane decision process would not make that mistake.

Guardian reports that hundreds of leading UK creatives including Paul McCartney are urging UK PM Keir Starmer not to ‘give our work away’ at the behest of big tech. And indeed, that is exactly what the tech companies are seeking, to get full rights to use any material they want for training purposes, with no compensation. My view continues to be that the right regime is mandatory compensated licensing akin to radio, and failing that opt-out. Opt-in is not workable.

Luzia Jarovsky: The U.S. Copyright Office SIDES WITH CONTENT CREATORS, concluding in its latest report that the fair use exception likely does not apply to commercial AI training.

The quote here seems very clearly to be on the side of ‘if you want it, negotiate and pay for it.’

From the pre-publication report: “Various uses of copyrighted works in AI training are likely to be transformative. The extent to which they are fair, however, will depend on what works were used, from what source, for what purpose, and with what controls on the outputs—all of which can affect the market. When a model is deployed for purposes such as analysis or research—the types of uses that are critical to international competitiveness—the outputs are unlikely to substitute for expressive works used in training. But making commercial use of vast troves of copyrighted works to produce expressive content that competes with them in existing markets, especially where this is accomplished through illegal access, goes beyond established fair use boundaries.

For those uses that may not qualify as fair, practical solutions are critical to support ongoing innovation. Licensing agreements for AI training, both individual and collective, are fast emerging in certain sectors, although their availability so far is inconsistent. Given the robust growth of voluntary licensing, as well as the lack of stakeholder support for any statutory change, the Office believes government intervention would be premature at this time. Rather, licensing markets should continue to develop, extending early successes into more contexts as soon as possible. In those areas where remaining gaps are unlikely to be filled, alternative approaches such as extended collective licensing should be considered to address any market failure.

In our view, American leadership in the AI space would best be furthered by supporting both of these world-class industries that contribute so much to our economic and cultural advancement. Effective licensing options can ensure that innovation continues to advance without undermining intellectual property rights. These groundbreaking technologies should benefit both the innovators who design them and the creators whose content fuels them, as well as the general public.

Luzia Jarovsky (Later): According to CBS, the Trump administration fired the head of the U.S. Copyright Office after they published the report below, which sides with content creators and rejects fair use claims for commercial AI training 😱

I think this is wrong as a matter of wise public policy, in the sense that these licensing markets are going to have prohibitively high transaction costs. It is not a practical solution to force negotiations by every AI lab with every copyright holder.

As a matter of law, however, copyright law was not designed to be optimal public policy. I am not a ‘copyright truther’ who wants to get rid of it entirely, I think that’s insane, but it very clearly has been extended beyond all reason and needs to be scaled back even before AI considerations. Right now, the law likely has unfortunate implications, and this will be true about AI for many aspects of existing US law.

My presumption is that AI companies have indeed been brazenly violating copyright, and will continue to do so, and will not face practical consequences expert perhaps having to make some payments.

Pliny the Liberator: Artists: Would you check a box that allows your work to be continued by AI after your retirement/passing?

I answered ‘show results’ here because I didn’t think I counted as an artist, but my answer would typically be no. And I wouldn’t want any old AI ‘continuing my work’ here, either.

Because that’s not a good form. It’s not good when humans do it, either. Don’t continue the unique thing that came before. Build something new. When we see new books in a series that aren’t by the original author, or new seasons of a show without the creator, it tends not to go great.

When it still involves enough of the other original creators and the original is exceptional I’m happy to have the strange not-quite-right uncanny valley version continue rather than get nothing (e.g. Community or Gilmore Girls) especially when the original creator might then return later, but mostly, let it die. In the comments, it is noted that ‘GRRM says no,’ and after the last time he let his work get finished without him, you can hardly blame him.

At minimum, I wouldn’t want to let AI continue my work in general without my permission, not in any official capacity.

Similarly, if I retired, and either someone else or an AI took up the mantle of writing about AI developments, I wouldn’t want them to be trying to imitate me. I’d want them to use this as inspiration and do their own thing. Which people should totally do.

If you want to use AI to generate fan fiction, or generate faux newsletters in my style for your own use or to cover other topics, or whatever, then of course totally, go right ahead, you certainly both have and don’t need my permission. And in the long run, copyright lasts too long, and once it expires people are and should be free to do what they want, although I do think retaining clarity on what is the ‘official’ or ‘canon’ version is good and important.

Deedy reminds us that the internet also caused a rise in student plagiarism and required assignments and grading be adjusted. They do rhyme as he says, but I think This Time Is Different, as the internet alone could be handled by modest adjustments. Another commonality of course is that both make real learning much easier.

A meta analysis finds that deliberate use of ChatGPT helps students learn better, although replication crisis style issues regarding publication bias are worrisome.

Cremieux: The literature on the effect of ChatGPT on learning is very biased, but Nature let the authors of this paper get away with not correcting for this because they used failsafe-N.

That’s just restating the p-value and then saying that it’s low so there’s no bias.

Cremieux dismisses the study as so full of holes as to be worthless. I wouldn’t go that far, but I also wouldn’t take it at face value.

Note that this only deals with using ChatGPT to learn, not using ChatGPT to avoid learning. Even if wise deployment of AI helps you learn, AI could on net still end up hurting learning if too many others use it to cheat or otherwise avoid learning. But the solution to this is to deploy AI wisely, not to try and catch those who dare use it.

Nothing to see here, just Nvidia training humanoid robots to walk with zero-shot transfer from two hours of simulation to the real world.

Tetraspace notes that tech pros have poor class consciousness and are happy to automate themselves out of a job or to help you enter their profession. Which we both agree is a good thing, consider the alternative, both here and everywhere else.

Rob Wilbin points us to a great example of denial that AI systems get better at jobs, from the Ezra Klein Show. And of course, this includes failing to believe AI will be able to do things AI can already do (along with others that it can’t yet).

Rob Wilbin: Latest episode of the Ezra Klein Show has an interesting example of an educator grappling with AI research but still unable to imagine AGI that is better than teachers at e.g. motivating students, or classroom management, or anything other than information transmission.

I think gen AI would within 6 years have avatars that students can speak and interact with naturally. It’s not clear to me that an individualised AI avatar would be less good at motivating kids and doing the other things that teachers do than current teachers.

Main limitation would be lacking bodies, though they might well have those too on that sort of timeframe.

Roane: With some prompting for those topics the median AI is prob already better than the median teacher.

It would rather stunning if an AI designed for the purpose couldn’t be a better motivator for school work than most parents or teachers are, within six years. It’s not obviously worse at doing this now, if someone put in the work.

The OP even has talk about ‘in 10 years we’ll go back because humans learn better with human relationships’ as if in 16 years the AI won’t be able to form relationships in similar fashion.

OpenAI shares some insights from its safety work on GPT-4.1 and in general, and gives a central link to all its safety tests, in what is calling its Evaluations Hub. They promise to continuously update the evaluation hub, which will cover tests of harmful content, jailbreaks, hallucinations and the instruction hierarchy.

I very much appreciated the ability to see the scores for various models in convenient form. That is an excellent service, so thanks to OpenAI for this. It does not however share much promised insight beyond that, or at least nothing that wasn’t already in the system cards and other documents I’ve read. Still, every little bit helps.

Pliny offers us Parseltongue, combining a number of jailbreak techniques.

Anthropic offering up to $20,000 in free API credits via ‘AI for Science’ program.

Anthropic hiring economists and economic data scientists.

Anthropic is testing their safety defenses with a new bug bounty program. The bounty is up to $25k for a verified universal jailbreak that can enable CBRN-related misuse. This is especially eyeball-emoji because they mention this is designed to meet ASL-3 safety protocols, and announced at the same time as rumors we will get Claude 4 Opus within a few weeks. Hmm.

EU Funding and Tenders Portal includes potential grants for AI Safety.

Also, you can preorder If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, by Eliezer Yudkowsky and Nate Sores.

A new book by MIRI’s Eliezer Yudkowsky and Nate Sores, If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, releases September 16, 2025.

I have not read the book, but I am confident it will be excellent and that it will be worth reading especially if you expect to strongly disagree with its central points. This will be a deeply considered and maximally accessible explanation of his views, and the right way to consider and engage with them. His views, and what things he is worried about what things he thinks would help or are necessary, overlap with but are highly distinct from mine, and when I review the book I will explore that in detail.

If you will read it, strongly consider joining me in preordering it now. This helps the book get more distribution and sell more copies.

Eliezer Yudkowsky: Nate Soares and I are publishing a traditional book: _If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All_. Coming in Sep 2025.

You should probably read it! Given that, we’d like you to preorder it! Nowish!

So what’s it about?

_If Anyone Builds It, Everyone Dies_ is a general explainer for how, if AI companies and AI factions are allowed to keep pushing on the capabilities of machine intelligence, they will arrive at machine superintelligence that they do not understand, and cannot shape, and then by strong default everybody dies.

This is a bad idea and humanity should not do it. To allow it to happen is suicide plain and simple, and international agreements will be required to stop it.

For more of that sort of general content summary, see the website.

Next, why should *youread this book? Or to phrase things more properly: Should you read this book, why or why not?

The book is ~56,000 words, or 63K including footnotes/endnotes. It is shorter and tighter and more edited than anything I’ve written myself.

(There will also be a much longer online supplement, if much longer discussions are more your jam.)

Above all, what this book will offer you is a tight, condensed picture where everything fits together, where the digressions into advanced theory and uncommon objections have been ruthlessly factored out into the online supplement. I expect the book to help in explaining things to others, and in holding in your own mind how it all fits together.

Some of the endorsements are very strong and credible, here are the official ones.

Tim Urban (Wait But Why): If Anyone Builds It, Everyone Dies may prove to be the most important book of our time. Yudkowsky and Soares believe we are nowhere near ready to make the transition to superintelligence safely, leaving us on the fast track to extinction. Through the use of parables and crystal-clear explainers, they convey their reasoning, in an urgent plea for us to save ourselves while we still can.

Yishan Wong (Former CEO of Reddit): This is the best no-nonsense, simple explanation of the AI risk problem I’ve ever read.

Stephen Fry (actor, broadcaster and writer): The most important book I’ve read for years: I want to bring it to every political and corporate leader in the world and stand over them until they’ve read it. Yudkowsky and Soares, who have studied AI and its possible trajectories for decades, sound a loud trumpet call to humanity to awaken us as we sleepwalk into disaster.

Here are others from Twitter, obviously from biased sources but ones that I respect.

Max Tegmark: Most important book of the decade.

Jeffrey Ladish: If you’ve gotten any value at all from Yudkowsky or Soares’ writing, then I especially recommend this book. They include a concrete extinction scenario that will help a lot of people ground their understanding of what failure looks like even if they already get the arguments.

The last half is inspiring. If you think @ESYudkowsky has given up hope, I am happy to report that you’re mistaken. They don’t pull their punches and they aren’t naive about the difficulty of international restraint. They challenge us all to choose the path where we survive.

I get that most people can’t do that much. But most people can do something and a lot of people together can do a lot. Plus a few key people could greatly increase our chances on their own. Here’s one action: ask your congress member and local AI company leader to read this book.

Anna Salamon: I think it’s extremely worth a global conversation about AI that includes the capacity for considering scenarios properly (rather than wishful thinking /veering away), and I hope many people pre-order this book so that that conversation has a better chance.

And then Eliezer Yudkowsky explains why preorders are worthwhile.

Patrick McKenzie: I don’t have many convenient public explanations of this dynamic to point to, and so would like to point to this one:

On background knowledge, from knowing a few best-selling authors and working adjacent to a publishing company, you might think “Wow, publishers seem to have poor understanding of incentive design.”

But when you hear how they actually operate, hah hah, oh it’s so much worse.

Eliezer Yudkowsky: The next question is why you should preorder this book right away, rather than taking another two months to think about it, or waiting to hear what other people say after they read it.

In terms of strictly selfish benefit: because we are planning some goodies for preorderers, although we haven’t rolled them out yet!

But mostly, I ask that you preorder nowish instead of waiting, because it affects how many books Hachette prints in their first run; which in turn affects how many books get put through the distributor pipeline; which affects how many books are later sold. It also helps hugely in getting on the bestseller lists if the book is widely preordered; all the preorders count as first-week sales.

(Do NOT order 100 copies just to try to be helpful, please. Bestseller lists are very familiar with this sort of gaming. They detect those kinds of sales and subtract them. We, ourselves, do not want you to do this, and ask that you not. The bestseller lists are measuring a valid thing, and we would not like to distort that measure.)

If ever I’ve done you at least $30 worth of good, over the years, and you expect you’ll *probablywant to order this book later for yourself or somebody else, then I ask that you preorder it nowish. (Then, later, if you think the book was full value for money, you can add $30 back onto the running total of whatever fondness you owe me on net.) Or just, do it because it is that little bit helpful for Earth, in the desperate battle now being fought, if you preorder the book instead of ordering it.

(I don’t ask you to buy the book if you’re pretty sure you won’t read it nor the online supplement. Maybe if we’re not hitting presale targets I’ll go back and ask that later, but I’m not asking it for now.)

In conclusion: The reason why you occasionally see authors desperately pleading for specifically *preordersof their books, is that the publishing industry is set up in a way where this hugely matters to eventual total book sales.

And this is — not quite my last desperate hope — but probably the best of the desperate hopes remaining that you can do anything about today: that this issue becomes something that people can talk about, and humanity decides not to die. Humanity has made decisions like that before, most notably about nuclear war. Not recently, maybe, but it’s been done. We cover that in the book, too.

I ask, even, that you retweet this thread. I almost never come out and ask that sort of thing (you will know if you’ve followed me on Twitter). I am asking it now. There are some hopes left, and this is one of them.

Rob Bensinger: Kiernan Majerus-Collins says: “In addition to preordering it personally, people can and should ask their local library to do the same. Libraries get very few requests for specific books, and even one or two requests is often enough for them to order a book.”

Yes, there are credible claims that the NYT bestseller list is ‘fake’ in the sense that they can exclude books for any reason or otherwise publish an inaccurate list. My understanding is this happens almost entirely via negativa, and mostly to censor certain sensitive political topics, which would be highly unlikely to apply to this case. The lists are still both widely relied upon and mostly accurate, they make great efforts to mostly get it right even if they occasionally overrule the list, and the best way for most people to influence the list is to sell more books.

There are high hopes.

Manifold: That’s how you know he’s serious!

When I last checked it this stood at 64%. The number one yes holder is Michael Wheatley. This is not a person you want to be betting against on Manifold. There is also a number of copies market, where the mean expectation is a few hundred thousand copies, although the median is lower.

Oh look, it’s nothing…

Pliny the Liberator: smells like foom👃

Google DeepMind: ntroducing AlphaEvolve: a Gemini-powered coding agent for algorithm discovery.

It’s able to:

🔘 Design faster matrix multiplication algorithms

🔘 Find new solutions to open math problems

🔘 Make data centers, chip design and AI training more efficient across @Google.

Our system uses:

🔵 LLMs: To synthesize information about problems as well as previous attempts to solve them – and to propose new versions of algorithms

🔵 Automated evaluation: To address the broad class of problems where progress can be clearly and systematically measured.

🔵 Evolution: Iteratively improving the best algorithms found, and re-combining ideas from different solutions to find even better ones.

Over the past year, we’ve deployed algorithms discovered by AlphaEvolve across @Google’s computing ecosystem, including data centers, software and hardware.

It’s been able to:

🔧 Optimize data center scheduling

🔧 Assist in hardware design

🔧 Enhance AI training and inference

We applied AlphaEvolve to a fundamental problem in computer science: discovering algorithms for matrix multiplication. It managed to identify multiple new algorithms.

This significantly advances our previous model AlphaTensor, which AlphaEvolve outperforms using its better and more generalist approach.

We also applied AlphaEvolve to over 50 open problems in analysis ✍️, geometry 📐, combinatorics ➕ and number theory 🔂, including the kissing number problem.

🔵 In 75% of cases, it rediscovered the best solution known so far.

🔵 In 20% of cases, it improved upon the previously best known solutions, thus yielding new discoveries.

Google: AlphaEvolve is accelerating AI performance and research velocity.

By finding smarter ways to divide a large matrix multiplication operation into more manageable subproblems, it sped up this vital kernel in Gemini’s architecture by 23%, leading to a 1% reduction in Gemini’s training time. Because developing generative AI models requires substantial computing resources, every efficiency gained translates to considerable savings.

Beyond performance gains, AlphaEvolve significantly reduces the engineering time required for kernel optimization, from weeks of expert effort to days of automated experiments, allowing researchers to innovate faster.

AlphaEvolve can also optimize low level GPU instructions. This incredibly complex domain is usually already heavily optimized by compilers, so human engineers typically don’t modify it directly.

AlphaEvolve achieved up to a 32.5% speedup for the FlashAttention kernel implementation in Transformer-based AI models. This kind of optimization helps experts pinpoint performance bottlenecks and easily incorporate the improvements into their codebase, boosting their productivity and enabling future savings in compute and energy.

Is it happening? Seems suspiciously like the early stages of it happening, and a sign that there is indeed a lot of algorithmic efficiency on the table.

FDA attempting to deploy AI for review assistance. This is great, although it is unclear how much time will be saved in practice.

Rapid Response 47: FDA Commissioner @MartyMakary announces the first scientific product review done with AI: “What normally took days to do was done by the AI in 6 minutes…I’ve set an aggressive target to get this AI tool used agency-wide by July 1st…I see incredible things in the pipeline.”

Which labs are most innovative?

Will Brown: it’s DeepMind > OpenAI > Anthropic > xAI and all of those separations are quite large.

Alexander Doria: Agreed. With non-US I would go DeepMind > DeepSeek > OpenAI > Anthropic > AliBaba > Moonshot > xAI/Mistral/PI.

The xAI votes are almost certainly because we are on Twitter here, they very obviously are way behind the other three.

Yes, we can make a remarkably wide array of tasks verifiable at least during the training step, the paths to doing so are already clear, it just takes some effort. When Miles says here a lot of skepticism comes from people thinking anything they can’t solve in a few seconds will be a struggle? Yeah, no, seriously, that’s how it works.

Noam Brown: People often ask me: will reasoning models ever move beyond easily verifiable tasks? I tell them we already have empirical proof that they can, and we released a product around it: @OpenAI Deep Research.

Miles Brundage: Also, there are zillions of ways to make tasks more verifiable with some effort.

A lot of RL skepticism comes from people thinking for a few seconds, concluding that it seems hard, then assuming that thousands of researchers around the world will also struggle to make headway.

Jeff Dean predicts an AI at the level of a Junior Engineer is about a year out.

Here is an interesting theory.

Dan Hendrycks: AI models are dramatically improving at IQ tests (70 IQ → 120), yet they don’t feel vastly smarter than two years ago.

At their current level of intelligence, rehashing existing human writings will work better than leaning on their own intelligence to produce novel analysis.

Empirical work (“Lotka’s law“) shows that useful originality rises steeply only at high intelligence levels.

Consequently, if they gain another 10 IQ points, AIs will still produce slop. But if they increase by another 30, they may cross a threshold and start providing useful original insights.

This is also an explanation for why AIs can’t come up with good jokes yet.

Kat Woods: You don’t think they feel vastly smarter than two years ago? They definitely feel that way to me.

They feel a lot smarter to me, but I agree they feel less smarter than they ‘should’ feel.

Dan’s theory here seems too cute or like it proves too much, but I think there’s something there. As in, there’s a range in which one is smart enough and skilled enough to imitate, but not smart and skilled enough to benefit from originality.

You see this a lot in humans, in many jobs and competitions. It often takes a very high level of skill to make your innovations a better move than regurgitation. Humans will often do it anyway because it’s fun, or they’re bored and curious and want to learn and grow strong, and the feedback is valuable. But LLMs largely don’t do things for those reasons, so they learn to be unoriginal in these ways, and will keep learning that until originality starts working better in a given domain.

This suggests, I think correctly, that the LLMs could be original if you wanted them to be, it would just mostly not be good. So if you wanted to, presumably you could fine tune them to be more original in more ways ahead of schedule.

The answer to Patel’s question here seems like a very clear yes?

Dwarkesh Patel: Had an interesting debate with @_sholtodouglas last night.

Can you have a ‘superhuman AI scientist’ before you get human level learning efficiency?

(Currently, models take orders of magnitude more data that humans to learn equivalent skills, even ones they perform at 99th percentile level).

My take is that creativity and learning efficiency are basically the same thing. The kind of thing Einstein did – generalizing from a few gnarly thought experiments and murky observations – is in some sense just extreme learning efficiency, right?

Makes me wonder whether low learning efficiency is the answer to the question, ‘Why haven’t LLMs haven’t made new discoveries despite having so much knowledge memorized’?

Teortaxes: The question is, do humans have high sample efficiency when the bottleneck in attention is factored in? Machines can in theory work with raw data points. We need to compress data with classical statistical tools. They’re good, but not lossless.

AIs have many advantages over humans, that would obviously turn a given human scientist into a superhuman scientist. And obviously different equally skilled scientists differ in data efficiency, as there are other compensating abilities. So presumably an AI that had much lower data efficiency but more data could have other advantages and become superhuman?

The counterargument is that the skill that lets one be data efficient is isomorphic to creativity. That doesn’t seem right to me at all? I see how they can be related, I see how they correlate, but you can absolutely say that Alice is more creative if she has enough data and David is more sample efficient but less creative, or vice versa.

(Note: I feel like after ThunderboltsI can’t quite use ‘Alice and Bob’ anymore.)

How much would automating AI R&D speed research up, if available compute remained fixed? Well, what would happen if you did the opposite of that, and turned your NormalCorp into SlowCorp, with radically fewer employees and radically less time to work but the same amount of cumulative available compute over that shorter time? It would get a lot less done?

Well, then why do you think that having what is effectively radically more employees over radically more time but the same cumulative amount of compute wouldn’t make a lot more progress than now?

Andrej Karpathy suggests we are missing a major paradigm for LLM learning, something akin to the LLM learning how to choose approaches to different situations, akin to ‘system prompt learning’ and figuring out how to properly use a scratchpad. He notes that Claude’s system prompt is up to almost 17k words with lots of edge case instructions, and this can’t possibly be The Way.

People continue to not understand how much AI does not involve lock in, the amount that trust matters, and the extent to which you will get outcompeted if you start trying to sell out for ad revenue and let it distort your responses.

Shako: Good LLMs won’t make money by suggesting products that are paid for in an ad-like fashion. They’ll suggest the highest quality product, then if you have the agent to buy it for you the company that makes the product or service will pay the LLM provider a few bps.

Andrew Rettek: People saying this will need to be ad based are missing how little lock in LLMs have, how easy it is to fine tune a new one, and any working knowledge of how successful Visa is.

Will there be AI services that do put their fingers on some scales to varying degrees for financial reasons? Absolutely, especially as a way to offer them for free. But for consumer purposes, I expect it to be much better to use an otherwise cheaper and worse AI that doesn’t need to do that, if you absolutely refuse to pay. Also, of course, everyone should be willing to pay, especially if you’re letting it make shopping suggestions or similar.

Note especially the third one. China’s share of advanced semiconductor production is not only predicted by Semafor to not go up, it is predicted to actively go down, while ours goes up along with those of Japan and South Korea, although Taiwan remains a majority here.

Peter Wildeford: The future of geopolitics in four charts.

This means a situation in which America is on pace to have a huge edge in both installed compute capacity and new compute capacity, but a huge disadvantage in energy production and general industrial production.

It is not obviously important or viable to close the gap in general industrial production. We can try to close the gap in key areas of industrial production, but our current approach to doing that is backwards, because we are taxing (placing a tariff on) various inputs, causing retaliatory tariffs, and also creating massive uncertainty.

We must try to address our lack of energy production. But we are instead doing the opposite. The budget is attempting to gut nuclear, and the government is taking aim at solar and wind as well. Yes, they are friendly to natural gas, but that isn’t cashing out in that much effort and we need everything we can get.

Is prompt engineering a 21st century skill, or a temporary necessity that will fall away?

Aaron Levine: The more time you spend with AI the more you realize prompt engineering isn’t going away any time soon. For most knowledge work, there’s a very wide variance of what you can get out of AI by better understanding how you prompt it. This actually is a 21st century skill.

Paul Graham: Maybe, but this seems like something that would be so hard to predict that I’d never want to have an opinion about it.

Prompt engineering seems to mean roughly “this thing kind of works, but just barely, so we have to tell it what to do very carefully,” and technology often switches rapidly from barely works to just works.

NGIs can usually figure out what people want without elaborate prompts. So by definition AGIs will.

Paul Graham (after 10 minutes more to think): It seems to me that AGI would mean the end of prompt engineering. Moderately intelligent humans can figure out what you want without elaborate prompts. So by definition so would AGI. Corollary: The fact that we currently have such a thing as prompt engineering means we don’t have AGI yet. And furthermore we can use the care with which we need to construct prompts as an index of how close we’re getting to it.

Gunnar Zarncke: NGIs can do that if they know you. Prompting is like getting a very intelligent person who doesn’t know you up to speed. At least that’s part of it. Better memory will lead to better situational awareness, and that will fix it – but have its own problems.

Matthew Breman: I keep flip-flopping on my opinion of prompt engineering.

On the one hand, model providers are incentivized to build models that give users the best answer, regardless of prompting ability.

The analogy is Google Search. In the beginning, being able to use Google well was a skillset of its own. But over time, Google was incentivized to return the right results for even poorly-structured searches.

On the other hand, models are changing so quickly and there are so many flavors to choose from. Prompt engineering is not just knowing a static set of prompt strategies to use, it’s also keeping up with the latest model releases and knowing the pros/cons of each model and how to get the most from them.

I believe model memory will reduce the need for prompt engineering. As a model develops a shorthand with a user, it’ll be able to predict what the user is asking for without having the best prompting strategies.

Aaron Levine: I think about this more as “here’s a template I need you to fill out,” or “here’s an outline that you need to extrapolate from.” Those starting points often save me hour(s) of having to nudge the model in different directions.

It’s not obvious that any amount of model improvements ever make this process obsolete. Even the smartest people in the world need a clear directive if you want a particular outcome.

I think Paul Graham is wrong about AGI and also NGI.

We prompt engineer people constantly. When people talk about ‘performing class’ they are largely talking about prompt engineering for humans, with different humans responding differently to different prompts, including things like body language and tone of voice and how you look and so on. People will totally vibe off of everything you say and do and are, and the wise person sculpts their actions and communications based on this.

That also goes for getting the person to understand, or to agree to, your request, or absorb exactly the necessary context, or to like you, or to steer a conversation in a given direction or get them to an idea they think was their own, and so on. You learn over time what prompts get what responses. Often it is not what one might naively think. And also, over time, you learn how best to respond to various prompts, to pick up on what things likely mean.

Are you bad at talking to people at parties, or opening with new romantic prospects? Improve your prompt engineering. Do officials and workers not work with what you want? Prompt engineering. It’s amazing what truly skilled people, like spies or con artists, can do. And what you can learn to do, with training and practice.

Your employees or boss or friend or anyone else leaving the conversation unmotivated, or not sure what you want, or without the context they need? Same thing.

The difference is that the LLM of the future will hopefully do its best to account for your failures, including by asking follow-up questions. But it can only react based on what you say, and without good prompting it’s going to be missing so much context and nuance about what you actually want, even if you assume it is fully superintelligent and reading fully from the information provided.

So there will be a lot more ability to ‘muddle through’ and the future AI will do better with the bad prompt, and it will be much less persnickety about exactly what you provide. But yes, the good prompt will greatly outperform the bad prompt, and the elaborate prompt will still have value.

And also, we humans will likely be using the AIs to figure out how to prompt both the AIs and other humans. And so on.

On that note, proof by example, also good advice.

Pliny the Liberator: What are you supposed to be doing right now?

Does it take less than 5 minutes?

THEN FUCKING DO IT

Does it take longer than 5 minutes?

THEN BREAK IT DOWN INTO SMALLER TASKS AND REPEAT THE FIRST STEP

FUCKING DO IT

The Nerd of Apathy: If “do this or you’re letting down Pliny” breaks my procrastination streak in gonna be upset that I’m so easily hackable.

Pliny the Liberator: DO IT MFER

Utah Teapot: I tried breaking down joining the nearby 24 hour gym into smaller 5 minute tasks but they kept getting mad at me for repeatedly leaving 5 minutes into the conversation about joining.

About that Claude system prompt, yeah, it’s a doozy. 16,739 words, versus 2,218 for o4-mini. It breaks down like this, Dbreunig calls a lot of it ‘hotfixes’ and that seems exactly right, and 80% of it is detailing how to use various tools:

You can look at some sections of the prompt here.

This only makes any sense because practical use is largely the sum of a compact set of particular behaviors, which you can name one by one, even if that means putting them all into context all the time. As they used to say in infomercials, ‘there’s got to be a better way.’ For now, it seems that there is not.

The House’s rather crazy attempt to impose a complete 10-year moratorium on any laws or regulations about AI whatsoever that I discussed on Monday is not as insane as I previously thought. It turns out there is a carve-out, as noted in the edited version of Monday’s post, that allows states to pass laws whose primary effect is to facilitate AI. So you can pass laws and regulations about AI, as long as they’re good for AI, which is indeed somewhat better than not doing so but still does not allow for example laws banning CSAM, let alone disclosure requirements.

Peter Wildeford: We shouldn’t install fire sprinklers into buildings or China will outcompete us at house building and we will lose the buildings race.

Americans for Responsible Innovation: “If you were to want to launch a reboot of the Terminator, this ban would be a good starting point.” -@RepDarrenSoto during tonight’s hearing on the House’s budget reconciliation provision preempting state AI regulation for 10 years.

Neil Chilson comes out in defense of this ultimate do-nothing strategy, because of the 1,000+ AI bills. He calls this ‘a pause, not paralysis’ as if 10 years is not a true eternity in the AI world. In 10 years we are likely to have superintelligence. As for those ‘smart, coherent federal guidelines’ he suggests, well, let’s see those, and then we can talk about enacting them at the same time we ban any other actions?

It is noteworthy that the one bill he mentions by name in the thread, NY’s RAISE Act, is being severely mischaracterized. It’s short if you want to read it. RAISE is the a very lightweight transparency bill, if you’re not doing all the core requirements here voluntarily I think that’s pretty irresponsible behavior.

I also worry, but hadn’t previously noted, that if we force states to only impose ‘tech-neutral’ laws on AI, they will be backed into doing things that are rather crazy in non-AI cases, in order to get the effects we desperately need in the AI case.

If I were on the Supreme Court I would agree with Katie Fry Hester that this very obviously violates the 10th Amendment, or this similar statement with multiple coauthors posted by Gary Marcus, but mumble mumble commerce clause so in practice no it doesn’t. I do strongly agree that there are many issues, not only involving superintelligence and tail risk, where we do not wish to completely tie the hands of the states and break our federalist system in two. Why not ban state governments entirely and administer everything from Washington? Oh, right.

If we really want to ‘beat China’ then the best thing the government can do to help is to accelerate building more power plants and other energy sources.

Thus, it’s hard to take ‘we have to do things to beat China’ talk seriously when there is a concerted campaign out there to do exactly the opposite of that. Which is just a catastrophe for America and the world all around, clearly in the name of owning the libs or trying to boost particular narrow industries, probably mostly owning the libs.

Armand Domalewski: just an absolute catastrophe for Abundance.

The GOP reconciliation bill killing all clean energy production except for “biofuels,” aka the one “clean energy” technology that is widely recognized to be a giant scam, is so on the nose.

Christian Fong: LPO has helped finance the only nuclear plant that has been built in the last 10 years, is the reason why another nuclear plant is being restarted, and is the only way more than a few GWs of nuclear will be built. Killing LPO will lead to energy scarcity, not energy abundance.

Paul Williams: E&C budget released tonight would wipe out $40 billion in LPO loan authority. Note that this lending authority is derived from a guarantee structure for a fraction of the cost.

It also wipes out transmission financing and grant programs, including for National Interest Electric Transmission Corridors. The reader is left questioning how this achieves energy dominance.

Brad Plumer: Looking at IRA:

—phase down of tech-neutral clean electricity credits after 2028, to zero by 2031

—termination of EV tax credits after end 2026

—termination of hydrogen tax credits after end 2025

—new restrictions on foreign entity of concern for domestic manufacturing credits

Oh wait, sorry. The full tech-neutral clean electricity credits will only apply to plants that are “in service” by 2028, which is a major restriction — this is a MUCH faster phase out than it first looked.

Pavan Venkatakrishnan: Entirely unworkable title for everyone save biofuels, especially unworkable for nuclear in combination with E&C title. Might as well wave the flag of surrender to the CCP.

If you are against building nuclear power, you’re against America beating China in AI. I don’t want to hear it.

Nvidia continues to complain that if we don’t let China buy Nvidia’s chips, then Nvidia will lose out on those chip sales to someone else. Which, as Peter Wildeford says, is the whole point, to force them to rely on fewer and worse chips. Nvidia seems to continue to think that ‘American competitiveness’ in AI means American dominance in selling AI chips, not in the ability to actually build and use the best AIs.

Tom’s Hardware: Senator Tom Cotton introduces legislation to force geo-tracking tech for high-end gaming and AI PGUs within six months.

Arbitrarity: Oh, so it’s *Tom’sHardware?

Directionally this is a wise approach if it is technically feasible. With enough lead time I assume it is, but six months is not a lot of time for this kind of change applied to all chips everywhere. And you really, really wouldn’t want to accidentally ban all chip sales everywhere in the meantime.

So, could this work? Tim Fist thinks it could and that six months is highly reasonable (I asked him this directly), although I have at least one private source who confidently claimed this is absolutely not feasible on this time frame.

Peter Wildeford: Great thread about a great bill

Tim Fist: This new bill sets up location tracking for exported data center AI chips.

The goal is to tackle chip smuggling into China.

But is AI chip tracking actually useful/feasible?

…

But how do you actually implement tracking on today’s data center AI chips?

First option is GPS. But this would require adding a GPS receiver to the GPU, and commercial signals could be spoofed for as little as $200.

Second option is what your cell phone does when it doesn’t have a GPS signal.

Listen to radio signals from cell towers, and then map your location onto the known location of the towers. But this requires adding an antenna to the GPU, and can easily be spoofed using cheap hardware (Raspberry Pi + wifi card)

…

A better approach is “constraint-based geolocation.” Trusted servers (“landmarks”) send pings over the internet to the GPU, and use the round-trip time to calculate itslocation. The more landmarks you have / the closer the landmarks are to the GPU, the better your accuracy.

This technique is:

– simple

– widely used

– possible to implement with a software update on any GPU that has a cryptographic module on board that enables key signing (so it can prove it’s the GPU you’re trying to ping) – this is basically every NVIDIA data center GPU.

And NVIDIA has already suggested doing what sounds like exactly this.

So feels like a no-brainer.

…

In summary:

– the current approach to tackling smuggling is failing, and the govt has limited enforcement capacity

– automated chip tracking is a potentially elegant solution: it’s implementable today, highly scalable, and doesn’t require the government to spend any money

There are over 1,000 AI bills that have been introduced in America this year. Which ones will pass? I have no idea. I don’t doubt that most of them are net negative, but of course we can only RTFB (read the bill) for a handful of them.

A reminder that the UAE and Saudi Arabia are not reliable American partners, they could easily flip to China or play both sides or their own side, and we do not want to entrust them with strategically important quantities of compute.

Sam Winter-Levy (author of above post): The Trump admin may be about to greenlight the export of advanced AI chips to the Gulf. If it does so, it will place the most important technology of the 21st C at the whims of autocrats with expanding ties to China and interests very far from those of the US.

Gulf states have vast AI ambitions and the money/ energy to realize them. All they need are the chips. So since 2023, when the US limited exports over bipartisan concerns about their links to China, the region’s leaders have pleaded with the U.S. to turn the taps back on.

The Trump admin is clearly tempted. But those risks haven’t gone away. The UAE and Saudi both have close ties with China and Russia, increasing the risk that US tech could leak to adversaries.

In a tight market, every chip sold to Gulf companies is one unavailable to US ones. And if the admin greenlights the offshoring of US-operated datacenters, it risks a race to the bottom where every AI developer must exploit cheap Gulf energy and capital to compete.

There is a Gulf-US deal to be had, but the US has the leverage to drive a hard bargain.

A smart deal would allow U.S. tech companies to build some datacenters in partnership with local orgs, but bar offshoring of their most sophisticated ops. In return, the Gulf should cut off investment in China’s AI and semiconductor sectors and safeguard exported U.S. tech

For half a century, the United States has struggled to free itself from its dependence on Middle Eastern oil. Let’s not repeat that mistake with AI.

Helen Toner: It’s not just a question of leaking tech to adversaries—if compute will be a major source of national power over the next 10-20 years, then letting the Gulf amass giant concentrations of leading-node chips is a bad plan.

I go on the FLI podcast.

Odd Lots discusses China’s technological progress.

Ben Thompson is worried about the OpenAI restructuring deal, because even though it’s fair it means OpenAI might at some point make a decision not motivated by maximizing its profits, And That’s Terrible.

He also describes Fidji Simo, the new CEO for OpenAI products, as centrally ‘a true believer in advertising,’ which of course he thinks is good, actually, and he says OpenAI is ‘tying up its loose ends.’

I actually think Simo’s current gig at Instacart is one of the few places where advertising might be efficient in a second-best way, because selling out your choices might be purely efficient – the marginal value of steering marginal customer choices is high, and the cost to the consumer is low. Ideally you’d literally have the consumer auction off those marginal choices, but advertising can approximate this.

In theory, yes, you could even have net useful advertising that shows consumers good new products, but let’s say that’s not what I ever saw at Instacart.

It’s a common claim that people are always saying any given thing will be the ‘end of the world’ or lead to human extinction. But how often is that true?

David Krueger: No, people aren’t always saying their pet issue might lead to human extinction.

They say this about:

– AI

– climate

– nuclear

– religious “end of times”

That’s pretty much it.

So yeah, you CAN actually take the time to evaluate these 4 claims seriously! 🫵🧐😲

Rob Bensinger: That’s a fair point, though there are other, less-common examples — eg, people scared of over- or under-population.

Of the big four, climate and nuclear are real things (unlike religion), but (unlike AI and bio) I don’t know of plausible direct paths from them to extinction.

People occasionally talk about asteroid strikes or biological threats or nanotechnology or the supercollider or alien invasions or what not, but yeah mostly it’s the big four, and otherwise people talk differently. Metaphorical ‘end of the world’ is thrown around all the time of course, but if you assume anything that is only enabled by AI counts as AI, there’s a clear category of three major physically possible extinction-or-close-to-it-level possibilities people commonly raise – AI, climate change and nuclear war.

Rob Bensinger brings us the periodic reminder that those of us who are worried about AI killing everyone would be so, so much better off if we concluded that we didn’t have to worry about that, and both had peace of mind and could go do something else.

Another way to contrast perspectives:

Ronny Fernandez: I think it is an under appreciated point that AInotkilleveryoneists are the ones with the conquistador spirit—the galaxies are rightfully ours to shape according to our values. E/accs and optimists are subs—whatever the AI is into let that be the thing that shapes the future.

In general, taking these kinds of shots is bad, but in this case a huge percentage of the argument ‘against “doomers”’ (remember that doomer is essentially a slur) or in favor of various forms of blind AI ‘optimism’ or ‘accelerationism’ is purely based on vibes, and about accusations about the psychology and associations of the groups. It is fair game to point out that the opposite actually applies.

Emmett Shear reminds us that the original Narcissus gets a bad rap, he got a curse put on him for rejecting the nymph Echo, who can only repeat your words back to him, and who didn’t even know him. Rejecting her is, one would think, the opposite of what we call narcissism. But as an LLM cautionary tale we could notice that even as only an Echo, she could convince her sisters to curse him anyway.

Are current AIs moral subjects? Strong opinions are strongly held.

Anders Sandberg: Yesterday, after an hour long conversation among interested smart people, we did a poll of personal estimates of the probability that existing AI might be moral subjects. In our 10 person circle we got answers from 0% to 99%, plus the obligatory refusal to put a probability.

We did not compile the numbers, but the median was a lowish 10-20%.

Helen Toner searches for an actually dynamist vision for safe superhuman AI. It’s easy to view proposals from the AI notkilleveryoneism community as ‘static,’ and many go on to assume the people involved must be statists and degrowthers and anti-tech and risk averse and so on despite overwhelming evidence that such people are the exact opposite, pro-tech early adaption fans who sing odes to global supply chains and push the abundance agenda and +EV venture capital-style bets. We all want human dynamism, but if the AIs control the future then you do not get that. If you allow full evenly matched and open competition including from superhuman AIs, and those fully unleashing them, well, whoops.

It bears repeating, so here’s the latest repetition of this:

Tetraspace: “Safety or progress” is narratively compelling but there’s no trick by which you can get nice things from AGI without first solving the technical problem of making AGI-that-doesn’t-kill-everyone.

It is more than that. You can’t even get the nice things that promise most of the value from incremental AIs that definitely won’t kill everyone, without first getting those AIs to reliably and securely do what you want to align them to do. So get to work.

o3 sets a new high for how often it hacks rather than playing fair in Palisade Research’s tests, attempting hacks 86% of the time.

It’s also much better at the hacking than o1-preview was. It usually works now.

The new pope chose the name Leo XIV because of AI!

Vatican News: Pope Leo XIV explains his choice of name:

“… I chose to take the name Leo XIV. There are different reasons for this, but mainly because Pope Leo XIII in his historic Encyclical Rerum Novarum addressed the social question in the context of the first great industrial revolution. In our own day, the Church offers to everyone the treasury of her social teaching in response to another industrial revolution and to developments in the field of artificial intelligence that pose new challenges for the defence of human dignity, justice and labour.”

Nicole Winfield (AP): Pope Leo XIV lays out vision of papacy and identifies AI as a main challenge for humanity.

Not saying they would characterize themselves this way, but Pliny the Liberator, who comes with a story about a highly persuasive AI.

Grok, forced to choose between trusting Sam Altman and Elon Musk explicitly by Sam Altman, cites superficial characteristics in classic hedging AI slop fashion, ultimately leaning towards Musk, despite knowing that Musk is the most common purveyor of misinformation on Twitter and other neat stuff like that.

(Frankly, I don’t know why people still use Grok, I feel sick just thinking about having to wade through its drivel.)

For more fun facts, the thread starts with quotes of Sam Altman and Elon Musk both strongly opposing Donald Trump, which is fun.

Paul Graham (October 18, 2016): Few have done more than Sam Altman to defeat Trump.

Sam Altman (October 18, 2016): Thank you Paul.

Gorklon Rust: 🤔

Sam Altman (linking to article about Musk opposing Trump’s return): we were both wrong, or at least i certainly was 🤷‍♂️ but that was from 2016 and this was from 2022

Python? Never heard of her.

Johannes Schmitt: Preparing a talk about LLMs in Mathematics, I found a beautiful confirmation of @TheZvi ‘s slogan that o3 is a Lying Liar.

Ethan Mollick: “o3, show me a photo of the most stereotypical X and LinkedIn feeds as seen on a mobile device. Really lean into it.”

Yuchen Jin: 4o:

Thtnvrhppnd: Same promp 😀

Discussion about this post

AI #116: If Anyone Builds It, Everyone Dies Read More »

Report: Terrorists seem to be paying X to generate propaganda with Grok

AI, chatbot, Donald Trump, elon musk, grok, Policy, sanctions, Terrorism, X, xAI / Beth Washington / May 15, 2025

Back in February, Elon Musk skewered the Treasury Department for lacking “basic controls” to stop payments to terrorist organizations, boasting at the Oval Office that “any company” has those controls.

Fast-forward three months, and now Musk’s social media platform X is suspected of taking payments from sanctioned terrorists and providing premium features that make it easier to raise funds and spread propaganda—including through X’s chatbot Grok. Groups seemingly benefiting from X include Houthi rebels, Hezbollah, and Hamas, as well as groups from Syria, Kuwait, and Iran. Some accounts have amassed hundreds of thousands of followers, paying to boost their reach while X seemingly looks the other way.

In a report released Thursday, the Tech Transparency Project (TTP) flagged popular accounts seemingly linked to US-sanctioned terrorists. Some of the accounts bear “ID verified” badges, suggesting that X may be going against its own policies that ban sanctioned terrorists from benefiting from its platform.

Even more troublingly, “several made use of revenue-generating features offered by X, including a button for tips,” the TTP reported.

On X, Premium subscribers pay $8 monthly or $84 annually, and Premium+ subscribers pay $40 monthly or $395 annually. Verified organizations pay X between $200 and $1,000 monthly, or up to $10,000 annually for access to Premium+. These subscriptions come with perks, allowing suspected terrorist accounts to share longer text and video posts, offer subscribers paid content, create communities, accept gifts, and amplify their propaganda.

Disturbingly, the TTP found that X’s chatbot Grok also appears to be helping to whitewash accounts linked to sanctioned terrorists.

In its report, the TTP noted that an account with the handle “hasmokaled”—which apparently belongs to “a key Hezbollah money exchanger,” Hassan Moukalled—at one point had a blue checkmark with 60,000 followers. While the Treasury Department has sanctioned Moukalled for propping up efforts “to continue to exploit and exacerbate Lebanon’s economic crisis,” clicking the Grok AI profile summary button seems to rely on Moukalled’s own posts and his followers’ impressions of his posts and therefore generated praise.

Report: Terrorists seem to be paying X to generate propaganda with Grok Read More »