Shannon Garcia – Page 17

Review: Mickey 17’s dark comedic antics make for a wild cinematic ride

bong joon-ho, culture, film reviews, Mickey 17, Robert Pattinson / Shannon Garcia / March 8, 2025

Mickey settles into his expendable role on the four-year journey, dying and being reprinted several times, and even finds love with security agent Nasha (Naomi Ackie). The mission finally reaches Niflheim, and he’s soon on Version 17—thanks to being used to detect a deadly airborne virus, with multiple versions dying in the quest to develop a vaccine. As the colonists explore this cold new world, Mickey 17 falls into a deep fissure inhabited by native life forms that resemble macroscale tardigrades, dubbed “creepers.” Timo leaves Mickey for dead, assuming they’ll just eat him, but the creepers (who seem to share a hive mind) instead save Mickey’s life, returning him to the surface.

Mickey Barnes (Robert Pattinson) failed to read the fine print when he signed up as an “expendable.” Warner Bros.

When Mickey gets back to his quarters, he finds his replacement, Mickey 18, is already there. The problem goes beyond Nasha’s opportunistic desire for an awkward threesome with the two Mickeys. Multiples are simply not allowed. The controversial reprinting technology isn’t even legal on Earth and was only allowed on the colonization mission with the understanding that any multiples would be killed immediately and their consciousness backup wiped—i.e., a permanent death.

A tale of two Mickeys

It’s Pattinson’s impressive dual performance as Mickey 17 and Mickey 18 that anchors the film. They might be clones with identical physical traits and memories, but we learn there are subtle differences in all the printings. Mickey 17 is more laid-back, meekly suffering abuse in the name of progress, while Mickey 18 is more rebellious and frankly has some anger issues. Pattinson adopted two different accents to differentiate between the two. Mickey and Nasha’s love story is the movie’s heart; she loves him in all his incarnations, through death after death. The scene where she dons a hazmat suit to hold Mickey 14—or is it 15?—in his isolation chamber as he dies (yet again) from the airborne virus is among the film’s most touching.

Review: Mickey 17’s dark comedic antics make for a wild cinematic ride Read More »

Nearly 1 million Windows devices targeted in advanced “malvertising” spree

Biz & IT, maltertising, malware, microsoft, Security / Shannon Garcia / March 8, 2025

A broad overview of the four stages. Credit: Microsoft

The campaign targeted “nearly” 1 million devices belonging both to individuals and a wide range of organizations and industries. The indiscriminate approach indicates the campaign was opportunistic, meaning it attempted to ensnare anyone, rather than targeting certain individuals, organizations, or industries. GitHub was the platform primarily used to host the malicious payload stages, but Discord and Dropbox were also used.

The malware located resources on the infected computer and sent them to the attacker’s c2 server. The exfiltrated data included the following browser files, which can store login cookies, passwords, browsing histories, and other sensitive data.

AppDataRoamingMozillaFirefoxProfiles.default-releasecookies.sqlite
AppDataRoamingMozillaFirefoxProfiles.default-releaseformhistory.sqlite
AppDataRoamingMozillaFirefoxProfiles.default-releasekey4.db
AppDataRoamingMozillaFirefoxProfiles.default-releaselogins.json
AppDataLocalGoogleChromeUser DataDefaultWeb Data
AppDataLocalGoogleChromeUser DataDefaultLogin Data
AppDataLocalMicrosoftEdgeUser DataDefaultLogin Data

Files stored on Microsoft’s OneDrive cloud service were also targeted. The malware also checked for the presence of cryptocurrency wallets including Ledger Live, Trezor Suite, KeepKey, BCVault, OneKey, and BitBox, “indicating potential financial data theft,” Microsoft said.

Microsoft said it suspects the sites hosting the malicious ads were streaming platforms providing unauthorized content. Two of the domains are movies7[.]net and 0123movie[.]art.

Microsoft Defender now detects the files used in the attack, and it’s likely other malware defense apps do the same. Anyone who thinks they may have been targeted can check indicators of compromise at the end of the Microsoft post. The post includes steps users can take to prevent falling prey to similar malvertising campaigns.

Nearly 1 million Windows devices targeted in advanced “malvertising” spree Read More »

Will the future of software development run on vibes?

AI, Biz & IT, coding, machine learning, Tech, vibe coding, vibecoding / Shannon Garcia / March 6, 2025

Accepting AI-written code without understanding how it works is growing in popularity.

For many people, coding is about telling a computer what to do and having the computer perform those precise actions repeatedly. With the rise of AI tools like ChatGPT, it’s now possible for someone to describe a program in English and have the AI model translate it into working code without ever understanding how the code works. Former OpenAI researcher Andrej Karpathy recently gave this practice a name—”vibe coding”—and it’s gaining traction in tech circles.

The technique, enabled by large language models (LLMs) from companies like OpenAI and Anthropic, has attracted attention for potentially lowering the barrier to entry for software creation. But questions remain about whether the approach can reliably produce code suitable for real-world applications, even as tools like Cursor Composer, GitHub Copilot, and Replit Agent make the process increasingly accessible to non-programmers.

Instead of being about control and precision, vibe coding is all about surrendering to the flow. On February 2, Karpathy introduced the term in a post on X, writing, “There’s a new kind of coding I call ‘vibe coding,’ where you fully give in to the vibes, embrace exponentials, and forget that the code even exists.” He described the process in deliberately casual terms: “I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.”

Karapthy tweet screenshot: There's a new kind of coding I call — A screenshot of Karpathy’s original X post about vibe coding from February 2, 2025. Credit: Andrej Karpathy / X

While vibe coding, if an error occurs, you feed it back into the AI model, accept the changes, hope it works, and repeat the process. Karpathy’s technique stands in stark contrast to traditional software development best practices, which typically emphasize careful planning, testing, and understanding of implementation details.

As Karpathy humorously acknowledged in his original post, the approach is for the ultimate lazy programmer experience: “I ask for the dumbest things, like ‘decrease the padding on the sidebar by half,’ because I’m too lazy to find it myself. I ‘Accept All’ always; I don’t read the diffs anymore.”

At its core, the technique transforms anyone with basic communication skills into a new type of natural language programmer—at least for simple projects. With AI models currently being held back by the amount of code an AI model can digest at once (context size), there tends to be an upper-limit to how complex a vibe-coded software project can get before the human at the wheel becomes a high-level project manager, manually assembling slices of AI-generated code into a larger architecture. But as technical limits expand with each generation of AI models, those limits may one day disappear.

Who are the vibe coders?

There’s no way to know exactly how many people are currently vibe coding their way through either hobby projects or development jobs, but Cursor reported 40,000 paying users in August 2024, and GitHub reported 1.3 million Copilot users just over a year ago (February 2024). While we can’t find user numbers for Replit Agent, the site claims 30 million users, with an unknown percentage using the site’s AI-powered coding agent.

One thing we do know: the approach has particularly gained traction online as a fun way of rapidly prototyping games. Microsoft’s Peter Yang recently demonstrated vibe coding in an X thread by building a simple 3D first-person shooter zombie game through conversational prompts fed into Cursor and Claude 3.7 Sonnet. Yang even used a speech-to-text app so he could verbally describe what he wanted to see and refine the prototype over time.

A photo of a MS-DOS computer with Q-BASIC code on the screen. — In August 2024, the author vibe coded his way into a working Q-BASIC utility script for MS-DOS, thanks to Claude Sonnet. Credit: Benj Edwards

We’ve been doing some vibe coding ourselves. Multiple Ars staffers have used AI assistants and coding tools for extracurricular hobby projects such as creating small games, crafting bespoke utilities, writing processing scripts, and more. Having a vibe-based code genie can come in handy in unexpected places: Last year, I asked Anthropic’s Claude write a Microsoft Q-BASIC program in MS-DOS that decompressed 200 ZIP files into custom directories, saving me many hours of manual typing work.

Debugging the vibes

With all this vibe coding going on, we had to turn to an expert for some input. Simon Willison, an independent software developer and AI researcher, offered a nuanced perspective on AI-assisted programming in an interview with Ars Technica. “I really enjoy vibe coding,” he said. “It’s a fun way to try out an idea and prove if it can work.”

But there are limits to how far Willison will go. “Vibe coding your way to a production codebase is clearly risky. Most of the work we do as software engineers involves evolving existing systems, where the quality and understandability of the underlying code is crucial.”

At some point, understanding at least some of the code is important because AI-generated code may include bugs, misunderstandings, and confabulations—for example, instances where the AI model generates references to nonexistent functions or libraries.

“Vibe coding is all fun and games until you have to vibe debug,” developer Ben South noted wryly on X, highlighting this fundamental issue.

Willison recently argued on his blog that encountering hallucinations with AI coding tools isn’t as detrimental as embedding false AI-generated information into a written report, because coding tools have built-in fact-checking: If there’s a confabulation, the code won’t work. This provides a natural boundary for vibe coding’s reliability—the code runs or it doesn’t.

Even so, the risk-reward calculation for vibe coding becomes far more complex in professional settings. While a solo developer might accept the trade-offs of vibe coding for personal projects, enterprise environments typically require code maintainability and reliability standards that vibe-coded solutions may struggle to meet. When code doesn’t work as expected, debugging requires understanding what the code is actually doing—precisely the knowledge that vibe coding tends to sidestep.

Programming without understanding

When it comes to defining what exactly constitutes vibe coding, Willison makes an important distinction: “If an LLM wrote every line of your code, but you’ve reviewed, tested, and understood it all, that’s not vibe coding in my book—that’s using an LLM as a typing assistant.” Vibe coding, in contrast, involves accepting code without fully understanding how it works.

While vibe coding originated with Karpathy as a playful term, it may encapsulate a real shift in how some developers approach programming tasks—prioritizing speed and experimentation over deep technical understanding. And to some people, that may be terrifying.

Willison emphasizes that developers need to take accountability for their code: “I firmly believe that as a developer you have to take accountability for the code you produce—if you’re going to put your name to it you need to be confident that you understand how and why it works—ideally to the point that you can explain it to somebody else.”

He also warns about a common path to technical debt: “For experiments and low-stake projects where you want to explore what’s possible and build fun prototypes? Go wild! But stay aware of the very real risk that a good enough prototype often faces pressure to get pushed to production.”

The future of programming jobs

So, is all this vibe coding going to cost human programmers their jobs? At its heart, programming has always been about telling a computer how to operate. The method of how we do that has changed over time, but there may always be people who are better at telling a computer precisely what to do than others—even in natural language. In some ways, those people may become the new “programmers.”

There was a point in the late 1970s to early ’80s when many people thought people required programming skills to use a computer effectively because there were very few pre-built applications for all the various computer platforms available. School systems worldwide made educational computer literacy efforts to teach people to code.

A brochure for the GE 210 computer from 1964. BASIC's creators used a similar computer four years later to develop the programming language. — A brochure for the GE 210 computer from 1964. BASIC’s creators used a similar computer four years later to develop the programming language that many children were taught at home and school. Credit: GE / Wikipedia

Before too long, people made useful software applications that let non-coders utilize computers easily—no programming required. Even so, programmers didn’t disappear—instead, they used applications to create better and more complex programs. Perhaps that will also happen with AI coding tools.

To use an analogy, computer controlled technologies like autopilot made reliable supersonic flight possible because they could handle aspects of flight that were too taxing for all but the most highly trained and capable humans to safely control. AI may do the same for programming, allowing humans to abstract away complexities that would otherwise take too much time to manually code, and that may allow for the creation of more complex and useful software experiences in the future.

But at that point, will humans still be able to understand or debug them? Maybe not. We may be completely dependent on AI tools, and some people no doubt find that a little scary or unwise.

Whether vibe coding lasts in the programming landscape or remains a prototyping technique will likely depend less on the capabilities of AI models and more on the willingness of organizations to accept risky trade-offs in code quality, maintainability, and technical debt. For now, vibe coding remains an apt descriptor of the messy, experimental relationship between AI and human developers—more collaborative than autonomous, but increasingly blurring the lines of who (or what) is really doing the programming.

Benj Edwards is Ars Technica’s Senior AI Reporter and founder of the site’s dedicated AI beat in 2022. He’s also a tech historian with almost two decades of experience. In his free time, he writes and records music, collects vintage computers, and enjoys nature. He lives in Raleigh, NC.

Will the future of software development run on vibes? Read More »

Andor S2 featurette teases canonical tragic event

Andor, culture, Disney Plus, Entertainment, star wars, streaming television / Shannon Garcia / March 6, 2025

Most of the main S1 cast is returning for S2, with the exception of Shaw. Forest Whitaker once again reprises his Rogue One role as Clone Wars veteran Saw Gerrera, joined by fellow Rogue One alums Ben Mendelsohn and Alan Tudyk as Orson Krennic and K-2SO, respectively. Benjamin Bratt has also been cast in an as-yet-undisclosed role.

The behind-the-scenes look opens with footage of a desperate emergency broadcast calling for help because Imperial ships were landing, filled with storm troopers intent on quashing any protesters or nascent rebels against the Empire who might be lurking about. “Revolutionary movements are spontaneously happening all over the galaxy,” series creator Tony Gilroy explains. “How those come together is the stuff of our story.” While S1 focused a great deal on political intrigue, Genevieve O’Reilly, who plays Mon Mothma, describes S2 as a “juggernaut,” with a size and scope to match.

The footage shown—some new, some shown in the last week’s teaser—confirms that assessment. There are glimpses of Gerrera, Krennic, and K-2SO, as well as Mothma’s home world, Chandrila. And are all those protesters chanting on the planet of Ghorman? That means we’re likely to see the infamous Ghorman Massacre, a brutal event that resulted in Mothma resigning from the Senate in protest against Emperor Palpatine. The massacre was so horrifying that it eventually served to mobilize and unite rebel forces across the galaxy in the Star Wars canon.

The first three (of 12) episodes of Andor S2 premiere on April 22, 2025, on Disney+. Subsequent three-episode chapters will drop weekly for the next three weeks after that.

poster art for Andor S2 — Credit: LucasFilm/Disney+

Andor S2 featurette teases canonical tragic event Read More »

Google tells Trump’s DOJ that forcing a Chrome sale would harm national security

antitrust, chrome, Google, national security, Policy / Shannon Garcia / March 6, 2025

Close-up of Google Chrome Web Browser web page on the web browser. Chrome is widely used web browser developed by Google. — Credit: Getty Images

The government’s 2024 request also sought to have Google’s investment in AI firms curtailed even though this isn’t directly related to search. If, like Google, you believe leadership in AI is important to the future of the world, limiting its investments could also affect national security. But in November, Mehta suggested he was open to considering AI remedies because “the recent emergence of AI products that are intended to mimic the functionality of search engines” is rapidly shifting the search market.

This perspective could be more likely to find supporters in the newly AI-obsessed US government with a rapidly changing Department of Justice. However, the DOJ has thus far opposed allowing AI firm Anthropic to participate in the case after it recently tried to intervene. Anthropic has received $3 billion worth of investments from Google, including $1 billion in January.

New year, new Justice Department

Google naturally opposed the government’s early remedy proposal, but this happened in November, months before the incoming Trump administration began remaking the DOJ. Since taking office, the new administration has routinely criticized the harsh treatment of US tech giants, taking aim at European Union laws like the Digital Markets Act, which tries to ensure user privacy and competition among so-called “gatekeeper” tech companies like Google.

We may get a better idea of how the DOJ wants to proceed later this week when both sides file their final proposals with Mehta. Google already announced its preferred remedy at the tail end of 2024. It’s unlikely Google’s final version will be any different, but everything is up in the air for the government.

Even if current political realities don’t affect the DOJ’s approach, the department’s staffing changes could. Many of the people handling Google’s case today are different than they were just a few months ago, so arguments that fell on deaf ears in 2024 could move the needle. Perhaps emphasizing the national security angle will resonate with the newly restaffed DOJ.

After both sides have had their say, it will be up to the judge to eventually rule on how Google must adapt its business. This remedy phase should get fully underway in April.

Google tells Trump’s DOJ that forcing a Chrome sale would harm national security Read More »

Threat posed by new VMware hyperjacking vulnerabilities is hard to overstate

Biz & IT, hypervsor, Security, Virtual Machine, vmware, vulnerabilities / Shannon Garcia / March 5, 2025

Three critical vulnerabilities in multiple virtual-machine products from VMware can give hackers unusually broad access to some of the most sensitive environments inside multiple customers’ networks, the company and outside researchers warned Tuesday.

The class of attack made possible by exploiting the vulnerabilities is known under several names, including hyperjacking, hypervisor attack, or virtual machine escape. Virtual machines often run inside hosting environments to prevent one customer from being able to access or control the resources of other customers. By breaking out of one customer’s isolated VM environment, a threat actor could take control of the hypervisor that apportions each VM. From there, the attacker could access the VMs of multiple customers, who often use these carefully controlled environments to host their internal networks.

All bets off

“If you can escape to the hypervisor you can access every system,” security researcher Kevin Beaumont said on Mastodon. “If you can escape to the hypervisor, all bets are off as a boundary is broken.” He added: “With this vuln you’d be able to use it to traverse VMware managed hosting providers, private clouds orgs have built on prem etc.”

VMware warned Tuesday that it has evidence suggesting the vulnerabilities are already under active exploitation in the wild. The company didn’t elaborate. Beaumont said the vulnerabilities affect “every supported (and unsupported)” version in VMware’s ESXi, Workstation, Fusion, Cloud Foundation, and Telco Cloud Platform product lines.

Threat posed by new VMware hyperjacking vulnerabilities is hard to overstate Read More »

Butch Wilmore says Elon Musk is “absolutely factual” on Dragon’s delayed return

crew-9, elon musk, NASA, Space / Shannon Garcia / March 5, 2025

For what it is worth, all of the reporting done by Ars over the last nine months suggests the decision to return Wilmore and Williams this spring was driven by technical reasons and NASA’s needs on board the International Space Station, rather than because of politics.

Q. How do you feel about waking up and finding yourself in a political storm?

Wilmore: I can tell you at the outset, all of us have the utmost respect for Mr. Musk, and obviously, respect and admiration for our president of the United States, Donald Trump. We appreciate them. We appreciate all that they do for us, for human space flight, for our nation. The words they said, politics, I mean, that’s part of life. We understand that. And there’s an important reason why we have a political system, a political system that we do have, and we’re behind it 100 percent. We know what we’ve lived up here, the ins and outs, and the specifics that they may not be privy to. And I’m sure that they have some issues that they are dealing with, information that they have, that we are not privy to. So when I think about your question, that’s part of life, we are on board with it.

Q. Did politics influence NASA’s decision for you to stay longer in space?

Wilmore: From my standpoint, politics is not playing into this at all. From our standpoint, I think that they would agree, we came up prepared to stay long, even though we plan to stay short. That’s what we do in human spaceflight. That’s what your nation’s human space flight program is all about, planning for unknown, unexpected contingencies. And we did that, and that’s why we flowed right into Crew 9, into Expedition 72 as we did. And it was somewhat of a seamless transition, because we had planned ahead for it, and we were prepared.

Butch Wilmore says Elon Musk is “absolutely factual” on Dragon’s delayed return Read More »

TSMC to invest $100B as Trump demands more US-made chips, report says

AI chips, Chips Act, chips fabs, Donald Trump, Policy, semiconductors, Taiwan Semiconductor Manufacturing Company, tariffs, TSMC / Shannon Garcia / March 4, 2025

Currently, TSMC only builds its most advanced chips in Taiwan. But when the most advanced US fabs are operational, they’ll be prepared to manufacture “tens of millions of leading-edge chips” to “power products like 5G/6G smartphones, autonomous vehicles, and AI datacenter servers,” the Commerce Department said in 2024.

TSMC has not confirmed the WSJ’s report but provided a statement: “We’re pleased to have an opportunity to meet with the President and look forward to discussing our shared vision for innovation and growth in the semiconductor industry, as well as exploring ways to bolster the technology sector along with our customers.”

Trump threat of semiconductor tariffs still looms

Advanced chips are regarded as critical for AI innovation, which Trump has prioritized, as well as for national security.

Without a steady supply, the US risks substantial technological and economic losses as well as potential weakening of its military.

To avert that, Trump campaigned on imposing tariffs that he claimed would drive more semiconductor manufacturing into the US, while criticizing the CHIPS Act for costing the US billions. Following through on that promise, in February, he threatened a “25 percent or more tariff” on all semiconductor imports, the WSJ reported. According to CNBC, Trump suggested those tariffs could be in effect by April 2.

“We have to have chips made in this country,” Trump said last month. “Right now, everything is made in Taiwan, practically, almost all of it, a little bit in South Korea, but everything—almost all of it is made in Taiwan. And we want it to be made—we want those companies to come to our country, in all due respect.”

While it’s unclear if Trump plans to overtly kill the CHIPS Act, his government funding cuts could trigger a future where the CHIPS Act dies with no workers left to certify that companies meet requirements for ongoing award disbursements, a semiconductor industry consultant group, Semiconductor Advisors, warned in a statement last month.

“If I were running a chip company, I would not count on CHIPS Act funding, even if I had a signed contract,” SA’s statement said.

TSMC to invest $100B as Trump demands more US-made chips, report says Read More »

Driving an EV restomod that costs as much as a house—the JIA Chieftain

Cars, Chieftain Range Rover, Range Rover, Restomod / Shannon Garcia / March 2, 2025

The Chieftain Range Rover is a fascinating thing—a refitted, reskinned, restored classic Range Rover is no new thing, nor is one with a ludicrous American V8 stuffed under the hood. But one that can be had as a gas car, plug-in hybrid, or as an EV? It can be all of those things depending on which boxes you tick. Ars Technica went for a spin in the EV to see how it stacks up.

The UK is something of an EV restomod hub. It’s been throwing electricity in things that didn’t come off the line electrified in the first place for years. Businesses like Electrogenic, Lunaz, and Everrati will, for a price, make an old car feel a little more peppy—depending on who you go to, it’ll come back restored as well. The Chieftain isn’t quite like them. Developed by Oxfordshire, UK, based Jensen International Automotive (the company’s bread ‘n butter is Jensen Interceptors), the Chieftain is an old Range Rover turned up to VERY LOUD. Or, actually, not loud at all.

Of course, these things come at a cost. A Chieftain EV Range Rover conversion, today, will set you back at least $568,000 should you choose to order one. This one was a private commission, and at that price there won’t be any built on spec on the off chance someone wants to buy one “off the peg.” By any stretch of the imagination it is a huge amount for an old car, but they’re custom-built from start to finish.

The Range Rover has aged well. Alex Goy

Yours will be made to your specification, have CarPlay/Android Auto, and the sort of mod cons one would expect in the 2020s. Under its perfectly painted shell—the color is your choice, of course—lives a 120 kWh battery. It’s made of packs mounted under the hood and in the rear, firing power to all four wheels via three motors: one at the front, and two at the rear. The tri-motor setup can theoretically produce around 650 hp (485 kW), but it’s paired back to a smidge over 405 hp (302 kW), so it doesn’t eat its tires on a spirited launch. There’s a 60: 40 rear-to-front torque split to keep things exciting if that’s your jam. Air suspension keeps occupants comfortable and insulated from the world around them.

Driving an EV restomod that costs as much as a house—the JIA Chieftain Read More »

Driving the new Mercedes CLA made me a believer in Mercedes-Benz’s EV future

Cars, Mercedes-Benz CLA, Mercedes-Benz Vision EQXX / Shannon Garcia / March 2, 2025

And if it doesn’t, they’ve got a hybrid version, too.

It’s not quite ready to be seen uncamouflaged, but Mercedes-Benz was ready to let us drive the new CLA. Credit: Tim Stevens

Mercedes-Benz’s EV efforts aren’t exactly burning up the sales charts. Models like the EQS and EQE haven’t convinced the brand’s demanding clientele that batteries are the future, forcing the company to scale back its electric ambitions.

Scale back, but not abandon. Benz is about to launch a new generation of EVs relying on technology derived from the epically efficient EQXX. The first is the new CLA. It’s coming soon, and after getting some time behind the wheel of a prototype vehicle undergoing final testing in the snowy wilds of Sweden, I’m convinced this could be the car to change Mercedes’ electrified fortunes.

And, for anyone who isn’t convinced, there’ll be a hybrid version too.

The EV is definitely the focus, though, and it tackles many of the most significant sticking points for would-be buyers of the company’s current electric offerings. First among those points is the styling. The somewhat anonymous shapes of the EQS and EQE have not earned those machines many fans outside of obsessive aerodynamicists. While the CLA I drove was unfortunately clad beneath some eye-warping camouflage, it seems to share enough lines with the Concept CLA Class that I’m already convinced it’ll be a looker.

The second concern with many of Benz’s current EVs is cost. Yes, you can get an EQB in the mid-$50,000 range, but that’s based on the older GLB. The least expensive of the company’s current EV range is the EQE sedan, with a mid-$70,000 starting price. That puts it well out of reach for many avid EV enthusiasts.

The front half. of a Mercedes-Benz CLA prototype — The CLA will have Mercedes’ first entirely in-house EV powertrain, and it’s far more efficient than the ones its currently offering. Credit: Tim Stevens

The current, gas-powered CLA starts in the mid-$40,000 range. Mercedes isn’t saying how much this new one will cost, but while the EV version will presumably be more, it should come in well beneath the EQE.

Next is the driving dynamic, which is really what brought me to Sweden. Both the EQS and EQE are fine cars, comfortable and calm with plenty of torque and power to be fun. However, they’re simply not the most engaging of machines. Can the CLA do better?

First impressions are definitely a yes. My driving was performed in the low-grip, wintery environment of northern Sweden, making it a little difficult to tell exactly how the car will feel when pushed in a more temperate world. But lowering the level of adhesion also lets you get an immediate sense of how well-balanced a machine is, and the CLA feels very well-balanced indeed.

When pushed beyond the limit of adhesion, it did have a tendency to understeer, but it didn’t take much provocation to bring the rear end around. Even with the stability control on, the 4matic-equipped car I drove was happy to swing out the rear as I danced from one corner to the next. When cruising at more relaxed speeds, the car soaked up the decidedly rough road surfaces extremely well for a car with such petite dimensions.

Most impressive was how well it handled the limited grip. One of the prime advantages of electrification is how quickly and smoothly stability and traction control can react to a loss of grip. The CLA didn’t immediately cut all power when it detected wheelspin, it quickly and automatically raised or lowered output to match the available grip.

The back half of a Mercedes CLA prototype — There will also be a hybrid version of the CLA for those who aren’t ready for a full EV. Credit: Tim Stevens

Power delivery, then, wasn’t all-or-nothing, and when it gave all it was plenty. The electric CLA felt comparably quick to the 402-hp EQE 500 4matic. The CLA 4matic makes similar power: 268 hp (200 kW) from the rear motor and 107 hp (80 kW) from the front. It gets off the line quickly, with the two-speed transmission on the rear axle ensuring that motor was still pulling strongly as I approached 100 mph (160 km/h).

Things were even more interesting when I needed to slow down. The CLA will be the debut of a new, unified braking system that effectively decouples the brake pedal from the actual physical action of the brakes. It’s not quite a full brake-by-wire system as there’s still a mechanical linkage there as a sort of fall-back, but in normal operation, the sensation you get from the brake pedal is entirely artificial, created by springs and bushings, not hydraulics.

There’s no feedback here, no haptics or adjustable resistance to signal what the brakes are doing. Indeed, the only indication that I’d triggered ABS on hard stops was the chatting noise coming from the wheels. In exchange, you get a consistent, solid brake feel, with the car mixing regenerative braking and physical braking as needed to deliver clean, consistent stops.

It’ll take more extensive testing to know how well the system handles something like a summer track day, but I can say that in my testing I got the brakes hot enough to be quite pungent, yet the car still stopped cleanly and predictably.

When it comes to one-pedal driving, the CLA offers a “D-” mode that will bring the car to a complete stop, but the outright deceleration rate after lifting off the accelerator is nowhere near as strong as something like a Tesla on Standard mode. That’s in addition to two lighter regen modes, plus “D Auto,” which varies regen based on surrounding traffic and terrain, just like the company’s current EVs.

A mercedes-benz CLA prototype seen head-on — The CLA was well-balanced on the ice. Credit: Tim Stevens

The CLA is also designed to address any concerns about efficiency with a number of improvements. That includes a new heat pump that can scavenge waste energy from the motor, the battery pack, and the ambient air. It’s said to heat the cabin twice as quickly with half the power consumption of the old heat pump.

There’s also a revised motor design, utilizing permanent magnets on both the front and rear axle. The system relies on a decoupling system to reduce drag on the front axle when it’s not needed, as on the EQE SUV, but the engagement is so quick and seamless that I never noticed.

The battery pack has also been revised, with a new chemistry that Mercedes says boosts overall energy density by 20 percent while also enabling a “significant reduction” in the use of cobalt.

The net result is a machine that promises to go 5.2 miles/kWh (11.9 kWh/100 km) and offers 466 miles (750 km) of range from the 85 kWh usable capacity “premium” battery pack. That’s on the European WLTP cycle, so on the American EPA cycle we can probably expect something closer to 400 miles (644 km). That still compares very favorably to the 308 miles (497 km) the current EQE can manage from its 96 kWh battery pack.

And, when you run out of juice, the new CLA’s 800-volt architecture enables charging rates of up to 320 kW. That theoretically means 186 miles (300 km) of charge in just 10 minutes.

The back of a Mercedes-Benz prototype in the snow — Battery energy density is up, and there’s a more efficient heat pump for the cabin. Credit: Tim Stevens

So, then, the promise is for a better-looking, better-driving, more-affordable, longer-range, and quicker-charging EV. That sounds like a winning bet, but Mercedes still has a hedge in. I didn’t just drive the electric CLA up in Sweden. I also got a go in the 48-volt hybrid version.

Yes, there’s a new CLA for you even if you’re still not on board with the EV revolution. This one’s built around a 1.5 L four-cylinder engine paired with an electric motor that’s integrated with an eight-speed, dual-clutch transmission. Engine output is rated at 188 hp (140 kW), plus an additional 27 hp (20 kW) from the electric motor.

That’s enough to drive the car up to 62 mph (100 kph) without spinning up the gasoline engine, but with only 1.3 kWh of battery at its disposal, you won’t go far without combustion. Mercedes doesn’t even quote an all-electric range. The engine comes on early and often.

In fact, during my time behind the wheel, I couldn’t get the engine to turn off. The engineers blamed the below-freezing temperatures. So, I can’t say just how sprightly the car will be without internal combustion. With that four-cylinder humming, the car was reasonably sprightly, the transmission slipping smoothly through the gears. Outright power is definitely on the limited side, though. Anyone who cares about acceleration should go for the other CLA, the one with the bigger battery.

The front of a Mercedes-Benz CLA prototype in the snow. — Mercedes-Benz may well have a winner here with the new CLA.

I got a good look at the interior of the two cars, but sadly, I’m not allowed to talk about that yet. Suffice it to say it includes some tasteful and practical changes that should be well-received. More on that to come.

Will the new CLA change Mercedes-Benz’s BEV fortunes? Initial impressions are indeed very good. If it looks half as good as that concept, delivers on the range promise, and is priced right, it should be a winner.

We won’t have long to wait to find out how it looks, but don’t expect an answer to the pricing question until closer to the car entering production later this year. Regardless, it’s great to see all the testing in the EQXX finally bearing fruit. At first bite, it’s tasting sweet.

Driving the new Mercedes CLA made me a believer in Mercedes-Benz’s EV future Read More »

We’ve figured out the basics of a shape-shifting, T-1000-style material

Science / Shannon Garcia / March 2, 2025

Campàs and his colleagues decided to design cell-like robots that could do all those things.

T-1000 building blocks

Each robot had motorized gears around its perimeter that could interlock with gears on other robots. The gears allowed the robots to move within the collective without breaking their bonds with each other, just like cells do in a living organism.

Linking the robots was a job of magnets that could rotate to maintain adhesion regardless of their orientation. Each robot also had a photodetector that could sense the polarity of light, allowing basic commands to be sent using a simple flashlight with a polarization filter. “The switch between solid and liquid states was driven by fluctuations of the force the motors applied, and we encoded the intensity of those fluctuations in the intensity of light,” says Matthew Devlin, a researcher at the Department of Mechanical Engineering at the University of California Santa Barbara and lead author of the study.

In response to light signals, two robotic collectives, 20 robots total, could elongate toward each other, touch in the middle, and form a bridge that could hold a load of just under 5 kilograms. After forming a cube, they could support an adult human weighing around 70 kilograms. They could also flow around an object, assume a complementary shape, and stiffen up to act as a wrench. “This was the Terminator idea of shapeshifting. This was exactly what we had in mind,” Campàs claims.

The only problem was, the robots were a bit above 5 centimeters in diameter. To get robotic collectives closer to Terminator’s mimetic polyalloy, the team wants to make the robots smaller. Much smaller.

Terminator nanobots?

“The good news is, you don’t have to go down with scale to what you see in living systems,” Campàs says. “Cells are roughly 10 microns. But anything around 100 microns—even up to 1 millimeter—robots would already be really impressive.” Unfortunately, we are rather far from making machines that small.

We’ve figured out the basics of a shape-shifting, T-1000-style material Read More »

SpaceX readies a redo of last month’s ill-fated Starship test flight

artemis, Commercial space, hls, launch, NASA, Science, Space, spacex, Starbase, starship, starship flight 8, super heavy, texas / Shannon Garcia / March 1, 2025

The FAA has cleared SpaceX to launch Starship’s eighth test flight as soon as Monday.

Ship 34, destined to launch on the next Starship test flight, test-fired its engines in South Texas on February 12. Credit: SpaceX

SpaceX plans to launch the eighth full-scale test flight of its enormous Starship rocket as soon as Monday after receiving regulatory approval from the Federal Aviation Administration.

The test flight will be a repeat of what SpaceX hoped to achieve on the previous Starship launch in January, when the rocket broke apart and showered debris over the Atlantic Ocean and Turks and Caicos Islands. The accident prevented SpaceX from completing many of the flight’s goals, such as testing Starship’s satellite deployment mechanism and new types of heat shield material.

Those things are high on the to-do list for Flight 8, set to lift off at 5: 30 pm CST (6: 30 pm EST; 23: 30 UTC) Monday from SpaceX’s Starbase launch facility on the Texas Gulf Coast. Over the weekend, SpaceX plans to mount the rocket’s Starship upper stage atop the Super Heavy booster already in position on the launch pad.

The fully stacked rocket will tower 404 feet (123.1 meters) tall. Like the test flight on January 16, this launch will use a second-generation, Block 2, version of Starship with larger propellant tanks with 25 percent more volume than previous vehicle iterations. The payload compartment near the ship’s top is somewhat smaller than the payload bay on Block 1 Starships.

This block upgrade moves SpaceX closer to attempting more challenging things with Starship, such as returning the ship, or upper stage, back to the launch site from orbit. It will be caught with the launch tower at Starbase, just like SpaceX accomplished last year with the Super Heavy booster. Officials also want to bring Starship into service to launch Starlink Internet satellites and demonstrate in-orbit refueling, an enabling capability for future Starship flights to the Moon and Mars.

NASA has contracts with SpaceX worth more than $4 billion to develop a Starship spinoff as a human-rated Moon lander for the Artemis lunar program. The mega-rocket is central to Elon Musk’s ambition to create a human settlement on Mars.

Another shot at glory

Other changes introduced on Starship Version 2 include redesigned forward flaps, which are smaller and closer to the tip of the ship’s nose to better protect them from the scorching heat of reentry. Technicians also removed some of the ship’s thermal protection tiles to “stress-test vulnerable areas” of the vehicle during descent. SpaceX is experimenting with metallic tile designs, including one with active cooling, that might be less brittle than the ceramic tiles used elsewhere on the ship.

Engineers also installed rudimentary catch fittings on the ship to evaluate how they respond to the heat of reentry, when temperatures outside the vehicle climb to 2,600° Fahrenheit (1,430° Celsius). Read more about Starship Version in this previous story from Ars.

It will take about 1 hour and 6 minutes for Starship to fly from the launch pad in South Texas to a splashdown zone in the Indian Ocean northwest of Australia. The rocket’s Super Heavy booster will fire 33 methane-fueled Raptor engines for two-and-a-half minutes as it climbs east from the Texas coastline, then jettison from the Starship upper stage and reverse course to return to Starbase for another catch with mechanical arms on the launch tower.

Meanwhile, Starship will ignite six Raptor engines and accelerate to a speed just shy of orbital velocity, putting the ship on a trajectory to reenter the atmosphere after soaring about halfway around the world.

Booster 15 perched on the launch mount at Starbase, Texas. Credit: SpaceX

If you’ve watched the last few Starship flights, this profile probably sounds familiar. SpaceX achieved successful splashdowns after three Starship test flights last year, and hoped to do it again before the premature end of Flight 7 in January. Instead, the accident was the most significant technical setback for the Starship program since the first full-scale test flight in 2023, which damaged the launch pad before the rocket spun out of control in the upper atmosphere.

Now, SpaceX hopes to get back on track. At the end of last year, company officials said they targeted as many as 25 Starship flights in 2025. Two months in, SpaceX is about to launch its second Starship of the year.

The breakup of Starship last month prevented SpaceX from evaluating the performance of the ship’s Pez-like satellite deployer and upgraded heat shield. Engineers are eager to see how those perform on Monday’s flight. Once in space, the ship will release four simulators replicating the approximate size and mass of SpaceX’s next-generation Starlink Internet satellites. They will follow the same suborbital trajectory as Starship and reenter the atmosphere over the Indian Ocean.

That will be followed by a restart of a Raptor engine on Starship in space, repeating a feat first achieved on Flight 6 in November. Officials want to ensure Raptor engines can reignite reliably in space before actually launching Starship into a stable orbit, where the ship must burn an engine to guide itself back into the atmosphere for a controlled reentry. With another suborbital flight on tap Monday, the engine relight is purely a confidence-building demonstration and not critical for a safe return to Earth.

The flight plan for Starship’s next launch includes another attempt to catch the Super Heavy booster with the launch tower, a satellite deployment demonstration, and an important test of its heat shield. Credit: SpaceX

Then, about 47 minutes into the mission, Starship will plunge back into the atmosphere. If this flight is like the previous few, expect to see live high-definition video streaming back from Starship as super-heated plasma envelops the vehicle in a cloak of pink and orange. Finally, air resistance will slow the ship below the speed of sound, and just 20 seconds before reaching the ocean, the rocket will flip to a vertical orientation and reignite its Raptor engines again to brake for splashdown.

This is where SpaceX hopes Starship Version 2 will shine. Although three Starships have made it to the ocean intact, the scorching temperatures of reentry damaged parts of their heat shields and flaps. That won’t do for SpaceX’s vision of rapidly reusing Starship with minimal or no refurbishment. Heat shield repairs slowed down the turnaround time between NASA’s space shuttle missions, and officials hope the upgraded heat shield on Starship Version 2 will decrease the downtime.

FAA’s green light

The FAA confirmed Friday it issued a launch license earlier this week for Starship Flight 8.

“The FAA determined SpaceX met all safety, environmental and other licensing requirements for the suborbital test flight,” an FAA spokesperson said in a statement.

The federal regulator oversaw a SpaceX-led investigation into the failure of Flight 7. SpaceX said NASA, the National Transportation Safety Board, and the US Space Force also participated in the investigation, which determined that propellant leaks and fires in an aft compartment, or attic, of Starship led to the shutdown of its engines and eventual breakup.

Engineers concluded the leaks were most likely caused by a harmonic response several times stronger than predicted, suggesting the vibrations during the ship’s climb into space were in resonance with the vehicle’s natural frequency. This would have intensified the vibrations beyond the levels engineers expected from ground testing.

Earlier this month, SpaceX completed an extended-duration static fire of the next Starship upper stage to test hardware modifications at multiple engine thrust levels. According to SpaceX, findings from the static fire informed changes to the fuel feed lines to Starship’s Raptor engines, adjustments to propellant temperatures, and a new operating thrust for the next test flight.

“To address flammability potential in the attic section on Starship, additional vents and a new purge system utilizing gaseous nitrogen are being added to the current generation of ships to make the area more robust to propellant leakage,” SpaceX said. “Future upgrades to Starship will introduce the Raptor 3 engine, reducing the attic volume and eliminating the majority of joints that can leak into this volume.”

FAA officials were apparently satisfied with all of this. The agency’s commercial spaceflight division completed a “comprehensive safety review” and determined Starship can return to flight operations while the investigation into the Flight 7 failure remains open. This isn’t new. The FAA also used this safety determination to expedite SpaceX launch license approvals last year as officials investigated mishaps on Starship and Falcon 9 rocket flights.

Stephen Clark is a space reporter at Ars Technica, covering private space companies and the world’s space agencies. Stephen writes about the nexus of technology, science, policy, and business on and off the planet.

SpaceX readies a redo of last month’s ill-fated Starship test flight Read More »

Author name: Shannon Garcia