Shannon Garcia – Page 73

$google-claims-math-breakthrough-with-proof-solving-ai-models$

Google claims math breakthrough with proof-solving AI models

AI, AlphaGeometry, AlphaGeometry 2, AlphaProof, Biz & IT, Google, google deepmind, machine learning / Shannon Garcia / July 27, 2024

slow and steady —

AlphaProof and AlphaGeometry 2 solve problems, with caveats on time and human assistance.

Benj Edwards – Jul 25, 2024 9: 54 pm UTC

Enlarge / An illustration provided by Google.

On Thursday, Google DeepMind announced that AI systems called AlphaProof and AlphaGeometry 2 reportedly solved four out of six problems from this year’s International Mathematical Olympiad (IMO), achieving a score equivalent to a silver medal. The tech giant claims this marks the first time an AI has reached this level of performance in the prestigious math competition—but as usual in AI, the claims aren’t as clear-cut as they seem.

Google says AlphaProof uses reinforcement learning to prove mathematical statements in the formal language called Lean. The system trains itself by generating and verifying millions of proofs, progressively tackling more difficult problems. Meanwhile, AlphaGeometry 2 is described as an upgraded version of Google’s previous geometry-solving AI modeI, now powered by a Gemini-based language model trained on significantly more data.

According to Google, prominent mathematicians Sir Timothy Gowers and Dr. Joseph Myers scored the AI model’s solutions using official IMO rules. The company reports its combined system earned 28 out of 42 possible points, just shy of the 29-point gold medal threshold. This included a perfect score on the competition’s hardest problem, which Google claims only five human contestants solved this year.

A math contest unlike any other

The IMO, held annually since 1959, pits elite pre-college mathematicians against exceptionally difficult problems in algebra, combinatorics, geometry, and number theory. Performance on IMO problems has become a recognized benchmark for assessing an AI system’s mathematical reasoning capabilities.

Google states that AlphaProof solved two algebra problems and one number theory problem, while AlphaGeometry 2 tackled the geometry question. The AI model reportedly failed to solve the two combinatorics problems. The company claims its systems solved one problem within minutes, while others took up to three days.

Google says it first translated the IMO problems into formal mathematical language for its AI model to process. This step differs from the official competition, where human contestants work directly with the problem statements during two 4.5-hour sessions.

Google reports that before this year’s competition, AlphaGeometry 2 could solve 83 percent of historical IMO geometry problems from the past 25 years, up from its predecessor’s 53 percent success rate. The company claims the new system solved this year’s geometry problem in 19 seconds after receiving the formalized version.

Limitations

Despite Google’s claims, Sir Timothy Gowers offered a more nuanced perspective on the Google DeepMind models in a thread posted on X. While acknowledging the achievement as “well beyond what automatic theorem provers could do before,” Gowers pointed out several key qualifications.

“The main qualification is that the program needed a lot longer than the human competitors—for some of the problems over 60 hours—and of course much faster processing speed than the poor old human brain,” Gowers wrote. “If the human competitors had been allowed that sort of time per problem they would undoubtedly have scored higher.”

Gowers also noted that humans manually translated the problems into the formal language Lean before the AI model began its work. He emphasized that while the AI performed the core mathematical reasoning, this “autoformalization” step was done by humans.

Regarding the broader implications for mathematical research, Gowers expressed uncertainty. “Are we close to the point where mathematicians are redundant? It’s hard to say. I would guess that we’re still a breakthrough or two short of that,” he wrote. He suggested that the system’s long processing times indicate it hasn’t “solved mathematics” but acknowledged that “there is clearly something interesting going on when it operates.”

Even with these limitations, Gowers speculated that such AI systems could become valuable research tools. “So we might be close to having a program that would enable mathematicians to get answers to a wide range of questions, provided those questions weren’t too difficult—the kind of thing one can do in a couple of hours. That would be massively useful as a research tool, even if it wasn’t itself capable of solving open problems.”

Google claims math breakthrough with proof-solving AI models Read More »

ISPs seeking government handouts try to avoid offering low-cost broadband

BEAD program, Policy / Shannon Garcia / July 26, 2024

But I don’t want to make broadband affordable —

Despite getting subsidies, ISPs oppose $30 plans for people with low incomes.

Jon Brodkin – Jul 26, 2024 6: 33 pm UTC

Illustration of fiber Internet cables — Getty Images | Yuichiro Chino

Internet service providers are eager to get money from a $42.45 billion government fund, but are trying to convince the Biden administration to drop demands that Internet service providers offer broadband service for as little as $30 a month to people with low incomes.

The Broadband Equity, Access, and Deployment (BEAD) program was created by a US law that requires Internet providers receiving federal funds to offer at least one “low-cost broadband service option for eligible subscribers.” The Biden administration says it is merely enforcing that legal requirement, but a July 23 letter sent by over 30 broadband industry trade groups claims that the administration is illegally regulating broadband prices.

The fund is administered by the National Telecommunications and Information Administration (NTIA). The NTIA is distributing money to states, which will then distribute it to ISPs. Before obtaining money from the NTIA, each state must get approval for a plan that includes a low-cost option. Nearly half of US states have already gotten approvals.

Although the law requires ISPs receiving grants to offer a low-cost plan, it also says the US may not “regulate the rates charged for broadband service.” In the letter sent to US Secretary of Commerce Gina Raimondo, ISPs claim that the NTIA’s demands for specific prices violate the ban on rate regulation:

We have also heard from stakeholders of specific instances in which certain State broadband offices have faced the prospect of political pressure unless they acceded to a $30 rate for the low-cost service option. This contravenes the clear language of the Infrastructure Act, which states that “[n]othing in this title may be construed to authorize [NTIA] to regulate the rates charged for broadband service.”

ISPs want to upend approved state plans

Funds like BEAD are intended to help ISPs build broadband networks in areas where it would otherwise not be economically feasible. In other words, the government giving money to ISPs directly lets the telcos make a decent profit on network-construction projects in areas where subscriber fees alone wouldn’t be enough.

ISPs receiving funds don’t have to offer the low-cost broadband plan to everyone. They only have to offer it to eligible subscribers who meet low-income requirements, as detailed in the NTIA’s Notice of Funding Opportunity.

Despite that, ISPs claim that prices for the low-cost option should be calculated based on “the economic realities of deploying and operating networks in the highest cost, hardest-to-reach areas.” The letter said:

While NTIA purports to give States the flexibility to choose a low-cost program that meets their particular needs, the reality is much different. According to NTIA’s own program guidance, it has “strongly encouraged” States to set a fixed rate of $30 per month for the low-cost service option. For a broad cross-section of America’s rural broadband providers, the $30 rate is completely unmoored from the economic realities of deploying and operating networks in the highest cost, hardest-to-reach areas that BEAD funding is precisely designed to reach.

Groups signing the letter include USTelecom, which represents AT&T, Verizon, CenturyLink/Lumen, and many other telcos. It was also signed by lobby groups for small cable firms and rural telcos, and numerous lobby groups for ISPs in specific states. The state-specific lobby groups signing the letter are from Alaska, Alabama, North Dakota, Montana, North Carolina, Kansas, Georgia, Illinois, Indiana, Iowa, Michigan, Minnesota, Nebraska, Nevada, New York, Ohio, Oregon, Oklahoma, Pennsylvania, South Carolina, South Dakota, Texas, Utah, Washington, and Wisconsin.

Many states have already received approval for their grant plans, including plans for requiring low-cost options. The NTIA today announced approval of New Mexico and Virginia’s initial proposals, bringing the total count to 22 states plus the Northern Mariana Islands, the District of Columbia, Puerto Rico, and the US Virgin Islands. Another 30 states and territories are waiting for approval after having submitted initial proposals by December 2023.

The lobby groups want the NTIA to reverse approvals for existing states’ plans. Their letter said the agency should “require each State to revise the low-cost service option rate proposed or approved in its Initial Proposal so that the rate is more reasonably tied to providers’ realistic costs, such as by using the FCC’s Urban Rate Survey benchmark.”

ISPs seeking government handouts try to avoid offering low-cost broadband Read More »

Barbie movie “may have spurred interest in gynecology,” study finds

Barbie, culture, gynecology, health, Popular Culture, Science, women's health / Shannon Garcia / July 26, 2024

Do you need a gynecologist? —

The movie apparently sparked some questions.

Beth Mole – Jul 26, 2024 5: 01 pm UTC

Enlarge / A digital advertisement board displaying a Barbie movie poster is seen in New York on July 24, 2023.

This post contains spoilers—for the movie and women’s health care.

There’s nothing like stirrups and a speculum to welcome one to womanhood, but for some, the recent Barbie movie apparently offered its own kind of eye-opening introduction.

The smash-hit film ends with the titular character making the brave decision to exit Barbieland and enter the real world as a bona fide woman. The film’s final scene follows her as she fully unfurls her new reality, attending her first woman’s health appointment. “I’m here to see my gynecologist,” she enthusiastically states to a medical receptionist. For many, the line prompted a wry chuckle, given her unsuspecting eagerness and enigmatic anatomy. But for others, it apparently raised some fundamental questions.

Online searches in the US for “gynecologist”—or alternate spellings, such as “gynaecologist”—rose an estimated 51 percent over baseline in the week following Barbie‘s July 21, 2023 release, according to a study published Thursday in JAMA Network Open. Moreover, searches related to the definition of gynecology spiked 154 percent. Those search terms included “gynecologist meaning,” “what is a gynecologist,” “what does a gynecologist do,” “why see a gynecologist,” and the weightiest of questions: “do I need a gynecologist.”

The “Barbie effect”

The study’s authors, led by researchers at Harvard Medical School, assessed 34 query terms that fit into six categories of searches, including “gynecologist,” “gynecologist definition,” “gynecologist appointment,” “doctors,” “doctor’s appointment,” and “women’s health.” The last three served as controls for more general interest in medical information. As the authors put it, the three control searches helped establish “whether unobserved contemporaneous factors influencing health-seeking behavior more generally may have contributed to gynecologic-related search volume.”

While the researchers noted clear spikes in “gynecologist” and “gynecologist definition” searches, they saw no changes in search trends for the three control search categories during the week after the movie’s release: “doctors,” “doctor’s appointment,” and “women’s health.” This suggested that the increase in gynecology-related searches may, in fact, be linked to the movie rather than some increased interest in health care generally.

The researchers also did not see a corresponding increase in searches related to gynecology appointments, suggesting that the transient online interest in gynecology didn’t translate to online searches for actual gynecological care, with queries such as “gynecologist near me.” The researchers speculate that two factors might explain this. For one, there’s the possibility that the data couldn’t capture care-seeking decisions. It may be that there’s a longer, variable time gap between newfound awareness of gynecology and the decision to seek care. The second possibility is that the people searching for basic information about gynecology may not need gynecological care themselves.

Overall, the authors conclude that “Barbie’s closing line may have spurred interest in gynecology.” The finding is supported by earlier work that also suggests popular culture can have measurable influences on health literacy and awareness among the general public, the authors conclude. For instance, when journalist Katie Couric live streamed her colonoscopy, there was a transient 21 percent increase in colonoscopies, and when actress Angelina Jolie penned an editorial about her experience with breast cancer, there was a transient 64 percent increase in genetic testing for breast cancer risk (BRCA testing). But while the “Barbie effect” seems to have raised some awareness of gynecology, it remains unclear if it will lead to a measurable improvement in health outcomes.

Barbie movie “may have spurred interest in gynecology,” study finds Read More »

At the Olympics, AI is watching you

2024 olympics, AI, AI surveillance, Biz & IT, Security, syndication / Shannon Garcia / July 26, 2024

“It’s the eyes of the police multiplied” —

New system foreshadows a future where there are too many CCTV cameras for humans to physically watch.

Morgan Meaker, WIRED.com – Jul 26, 2024 4: 24 pm UTC

Police observe the Eiffel Tower from Trocadero ahead of the Paris 2024 Olympic Games. — Enlarge / Police observe the Eiffel Tower from Trocadero ahead of the Paris 2024 Olympic Games on July 22, 2024.

On the eve of the Olympics opening ceremony, Paris is a city swamped in security. Forty thousand barriers divide the French capital. Packs of police officers wearing stab vests patrol pretty, cobbled streets. The river Seine is out of bounds to anyone who has not already been vetted and issued a personal QR code. Khaki-clad soldiers, present since the 2015 terrorist attacks, linger near a canal-side boulangerie, wearing berets and clutching large guns to their chests.

French interior minister Gérald Darmanin has spent the past week justifying these measures as vigilance—not overkill. France is facing the “biggest security challenge any country has ever had to organize in a time of peace,” he told reporters on Tuesday. In an interview with weekly newspaper Le Journal du Dimanche, he explained that “potentially dangerous individuals” have been caught applying to work or volunteer at the Olympics, including 257 radical Islamists, 181 members of the far left, and 95 from the far right. Yesterday, he told French news broadcaster BFM that a Russian citizen had been arrested on suspicion of plotting “large scale” acts of “destabilization” during the Games.

Parisians are still grumbling about road closures and bike lanes that abruptly end without warning, while human rights groups are denouncing “unacceptable risks to fundamental rights.” For the Games, this is nothing new. Complaints about dystopian security are almost an Olympics tradition. Previous iterations have been characterized as Lockdown London, Fortress Tokyo, and the “arms race” in Rio. This time, it is the least-visible security measures that have emerged as some of the most controversial. Security measures in Paris have been turbocharged by a new type of AI, as the city enables controversial algorithms to crawl CCTV footage of transport stations looking for threats. The system was first tested in Paris back in March at two Depeche Mode concerts.

For critics and supporters alike, algorithmic oversight of CCTV footage offers a glimpse of the security systems of the future, where there is simply too much surveillance footage for human operators to physically watch. “The software is an extension of the police,” says Noémie Levain, a member of the activist group La Quadrature du Net, which opposes AI surveillance. “It’s the eyes of the police multiplied.”

Near the entrance of the Porte de Pantin metro station, surveillance cameras are bolted to the ceiling, encased in an easily overlooked gray metal box. A small sign is pinned to the wall above the bin, informing anyone willing to stop and read that they are part of a “video surveillance analysis experiment.” The company which runs the Paris metro RATP “is likely” to use “automated analysis in real time” of the CCTV images “in which you can appear,” the sign explains to the oblivious passengers rushing past. The experiment, it says, runs until March 2025.

Porte de Pantin is on the edge of the park La Villette, home to the Olympics’ Park of Nations, where fans can eat or drink in pavilions dedicated to 15 different countries. The Metro stop is also one of 46 train and metro stations where the CCTV algorithms will be deployed during the Olympics, according to an announcement by the Prefecture du Paris, a unit of the interior ministry. City representatives did not reply to WIRED’s questions on whether there are plans to use AI surveillance outside the transport network. Under a March 2023 law, algorithms are allowed to search CCTV footage in real-time for eight “events,” including crowd surges, abnormally large groups of people, abandoned objects, weapons, or a person falling to the ground.

“What we’re doing is transforming CCTV cameras into a powerful monitoring tool,” says Matthias Houllier, cofounder of Wintics, one of four French companies that won contracts to have their algorithms deployed at the Olympics. “With thousands of cameras, it’s impossible for police officers [to react to every camera].”

At the Olympics, AI is watching you Read More »

Scientists unlock more secrets of Rembrandt’s pigments in The Night Watch

art conservation, chemistry, Physics, pigments, rembrandt, Science, The Night Watch / Shannon Garcia / July 25, 2024

Use of arsenic sulfides for yellow, orange/red hues adds to artist’s known pigment palette.

Jennifer Ouellette – Jul 25, 2024 7: 32 pm UTC

The Nightwatch, or Militia Company of District II under the Command of Captain Frans Banninck Cocq (1642) — Enlarge / Rembrandt’s *The Night Watch* underwent many chemical and mechanical alterations over the last 400 years.

Public domain

Since 2019, researchers have been analyzing the chemical composition of the materials used to create Rembrandt’s masterpiece, The Night Watch, as part of the Rijksmuseum’s ongoing Operation Night Watch, devoted to its long-term preservation. Chemists at the Rijksmuseum and the University of Amsterdam have now detected unusual arsenic-based yellow and orange/red pigments used to paint the duff coat of one of the central figures in the painting, according to a recent paper in the journal Heritage Science. It’s a new addition to Rembrandt’s known pigment palette that further adds to our growing body of knowledge about the materials he used.

As previously reported, past analyses of Rembrandt’s paintings identified many pigments the Dutch master used in his work, including lead white, multiple ochres, bone black, vermilion, madder lake, azurite, ultramarine, yellow lake, and lead-tin yellow, among others. The artist rarely used pure blue or green pigments, with Belshazzar’s Feast being a notable exception. (The Rembrandt Database is the best resource for a comprehensive chronicling of the many different investigative reports.)

Early last year, the researchers at Operation Night Watch found rare traces of a compound called lead formate in the painting—surprising in itself, but the team also identified those formates in areas where there was no lead pigment, white or yellow. It’s possible that lead formates disappear fairly quickly, which could explain why they have not been detected in paintings by the Dutch Masters until now. But if that is the case, why didn’t the lead formate disappear in The Night Watch? And where did it come from in the first place?

Hoping to answer these questions, the team whipped up a model of “cooked oils” from a 17th-century recipe and analyzed those model oils with synchrotron radiation. The results supported their hypothesis that the oil used for light parts of the painting was treated with an alkaline lead drier. The fact that The Night Watch was revarnished with an oil-based varnish in the 18th century complicates matters, as this may have provided a fresh source of formic acid, such that different regions of the painting rich in lead formates may have formed at different times in the painting’s history.

Last December, the team turned its attention to the preparatory layers applied to the canvas. It’s known that Rembrandt used a quartz-clay ground for The Night Watch—the first time he had done so, perhaps because the colossal size of the painting “motivated him to look for a cheaper, less heavy and more flexible alternative for the ground layer” than the red earth, lead white, and cerussite he was known to use on earlier paintings.

The Night Watch. (b) Detail of figure’s embroidered gold buff coat. (c) X-ray diffraction image of coat detail showing arsenic. (d) Stereomicroscope image showing arsenic hot spot.” height=”531″ src=”https://cdn.arstechnica.net/wp-content/uploads/2024/07/rembrandt1-640×531.jpg” width=”640″>

Enlarge / (a) Rembrandt’s The Night Watch. (b) Detail of figure’s embroidered gold buff coat. (c) X-ray diffraction image of coat detail showing arsenic. (d) Stereomicroscope image showing arsenic hot spot.

N. De Keyser et al., 2024

They used 3D X-ray methods to capture more detail, revealing the presence of an unknown (and unexpected) lead-containing layer located just underneath the ground layer. This could be due to using a lead compound added to the oil used to prepare the canvas as a drying additive—perhaps to protect the painting from the damaging effects of humidity. (Usually a glue sizing was used before applying the ground layer.) The lead layer discovered last year could be the reason for the unusual lead protrusions in areas of The Night Watch, since there are no other lead-containing compounds in the paint. It’s possible that lead migrated into the painting’s ground layer from the lead-oil preparatory layer below.

An intentional combination

The presence of arsenic sulfides in The Night Watch appears to be an intentional pigment combination by Rembrandt, according to the authors of this latest paper. Artists throughout history have used naturally occurring orpiment and realgar, as well as artificial arsenic sulfide pigments, to get yellow, orange, and red hues in their paints. Orpiment was also used for medicinal purposes, in hair removal creams and oils, in wax seals, yellow ink, bookbinder green (mixed with indigo), and for the treatment or coating of metals like silver.

However, the use of artificial arsenic sulfides has rarely been reported in artworks, although they are mentioned in multiple artists’ treatises dating back to the 15th century. Earlier work using advanced analytical techniques such as Raman spectroscopy and X-ray powder diffraction revealed that Rembrandt used arsenic sulfide pigments (artificial orpiment) in two late paintings: The Jewish Bride (c 1665) and The Man in a Red Cap (c 1665).

For this latest work, Nouchka De Keyser of the Rijksmuseum and co-authors used macroscopic X-ray fluorescence imaging to map The Night Watch, which revealed the presence of arsenic and sulfur in the doublet sleeves and embroidered buff coat worn by Lt. Willem Van Ruytenburch, i.e., the central figure to the right of Captain Frans Bannick Cocq in the painting. The researchers initially assumed that this was due to Rembrandt’s use of orpiment for yellow hues and realgar for red hues.

Ars Vitraria Experimentalis, 1679. (c) Page from the Weimar taxa of 1674 including prices for white, yellow, and red arsenic.” height=”300″ src=”https://cdn.arstechnica.net/wp-content/uploads/2024/07/rembrandt2-640×300.jpg” width=”640″>

Enlarge / (a, b) Pages from Johann Kunckel’s Ars Vitraria Experimentalis, 1679. (c) Page from the Weimar taxa of 1674 including prices for white, yellow, and red arsenic.

N. De Keyser et al., 2024

To learn more, they took tiny samples and analyzed them with light microscopy, micro-Raman spectroscopy, electron microscopy, and X-ray powder diffraction. They found the yellow particles were actually pararealgar while the orange to red particles were semi-amorphous pararealgar. These are more unusual arsenic sulfide components, typically associated with degradation products from either the natural minerals or their artificial equivalents as they age.

But De Keyser et al. concluded that the presence of these components was actually an intentional mixture, based on their perusal of multiple historical sources and catalogs of collection cabinets with long lists of various arsenic sulfides. There was clearly contemporary knowledge of manipulating both natural and artificial arsenic sulfides to get different shades of yellow, orange, and red.

They also found vermilion and lead-tin yellow in the paint mixture; Rembrandt was known to use these to add brightness and intensity to his paintings. In the case of The Night Watch, “Rembrandt clearly aimed for a bright orange tone with a high color strength that allowed him to create an illusion of the gold thread embroidery in Van Ruytenburch’s costume,” the authors wrote. “The artificial orange to red arsenic sulfide might have offered different optical and rheological paint properties as compared to the mineral form of orpiment and realgar.”

In addition, the team examined paint samples from different artists known to use arsenic sulfides—whose works are also part of the Rijksmuseum collection—and found a similar mixture of pigments in a painting by Rembrandt’s contemporary, Willem Kalf. “It is evidence that a variety of natural and artificial arsenic sulfides were manufactured and traded during Rembrandt’s time and were available in Amsterdam,” the authors wrote—most likely imported, since the Dutch Republic did not have considerable mining resources.

Heritage Science, 2024. DOI: 10.1186/s40494-024-01350-x (About DOIs).

Scientists unlock more secrets of Rembrandt’s pigments in The Night Watch Read More »

No, NASA hasn’t found life on Mars yet, but the latest discovery is intriguing

Life on Mars, Mars, perseverance, Science, Space / Shannon Garcia / July 25, 2024

Look at the big brain on percy —

“These spots are a big surprise.”

Eric Berger – Jul 25, 2024 6: 44 pm UTC

Enlarge / NASA’s Perseverance rover discovered “leopard spots” on a reddish rock nicknamed “Cheyava Falls” in Mars’ Jezero Crater in July 2024.

NASA/JPL-Caltech/MSSS

NASA’s Perseverance rover has found a very intriguing rock on the surface of Mars.

An arrowhead-shaped rock observed by the rover has chemical signatures and structures that could have been formed by ancient microbial life. To be absolutely clear, this is not irrefutable evidence of past life on Mars, when the red planet was more amenable to water-based life billions of years ago. But discovering these colored spots on this rock is darn intriguing and has Mars scientists bubbling with excitement.

“These spots are a big surprise,” said David Flannery, an astrobiologist and member of the Perseverance science team from the Queensland University of Technology in Australia, in a NASA news release. “On Earth, these types of features in rocks are often associated with the fossilized record of microbes living in the subsurface.”

What the rover found

This is a very recent discovery, and the science has not yet been peer-reviewed. The sample was collected on July 21—a mere four days ago—as the rover explored the Neretva Vallis riverbed. This valley was formed long ago when water rushed into Jezero Crater.

The science team operating Perseverance has nicknamed the rock Chevaya Falls and subjected it to multiple scans by the rover’s SHERLOC (Scanning Habitable Environments with Raman & Luminescence for Organics & Chemicals) instrument. The distinctive colorful spots, containing both iron and phosphate, are a smoking gun for certain chemical reactions—rather than microbial life itself.

On Earth, microbial life can derive energy from these kinds of chemical reactions. So, what we have here is a plausible source of energy for microbes on Mars. In addition, there are organic chemicals present on the same rock, which is consistent with something living there. From this, it is tempting to jump to the idea of microbes living on a rock, eons ago, in a Martian river. But this is not direct evidence of life.

NASA has a seven-step process for determining whether something can be confirmed as extraterrestrial life. This is known as the CoLD scale, for Confidence of Life Detection. In this case, the detection of these spots on a Martian rock represents just the first of seven steps—for example, scientists must still rule out non-biological possibility and identify other signals to have confidence in off-world life.

Bring them home

According to NASA, Perseverance has used all of its available instrumentation to study Chevaya Falls. “We have zapped that rock with lasers and X-rays and imaged it literally day and night from just about every angle imaginable,” said Ken Farley, Perseverance project scientist. “Scientifically, Perseverance has nothing more to give.”

The discovery provides some wind in the sails for NASA’s flagging efforts to devise and fly a Mars Sample Return mission. The agency’s most recent plan, costing $11 billion, was determined to be too expensive. Now, the space agency is asking the industry for help. In June it commissioned 10 studies on alternative means of returning rocks from Mars sooner, and presumably for a lower cost.

Now, scientists can point to rocks like Chevaya Falls and say this is precisely why they must be studied in ultra-capable labs back on Earth.

No, NASA hasn’t found life on Mars yet, but the latest discovery is intriguing Read More »

Ars is seeking a seasoned senior reporter for all things Google

android, ars, Feature, Google, hiring, jobs at ars, Staff, Tech / Shannon Garcia / July 25, 2024

get your ron on —

Got feelings about the future of AI and/or phone bezel width? Come apply!

Ars Staff – Jul 25, 2024 6: 00 pm UTC

A photograph of — Enlarge / If you get hired for this position, you’ll be provided an assistant. It’s this guy. This guy is your assistant. His name is “Googly.”

Google is a company in transformation—but “from what“ and “to what“ are not always clear. To catalog and examine Google’s moves in this new era of generative AI, Ars Technica is hiring a Senior Technology Reporter to focus on Google, AI, Android, and search. While attention to so-called “consumer products” will be important, this role will be more focused on Google’s big moves as a technology and infrastructure company, moves often made to counter perceived threats from companies like OpenAI, Microsoft, and Perplexity. Informed skepticism is the rule around here, so we’re looking for someone with the chops to bring a critical eye to some deep technical and business issues.

As this is a senior role owning an important beat, it is not an entry-level position. We’re looking for someone who can primarily self-direct when it comes to their reporting and someone who is comfortable working remotely within a similarly remote team. We’d also like someone who can bring to the table deep and intelligent analyses on broader Google topics while also hitting smaller daily news stories.

This is a full-time union job with benefits.

All candidates:

Must have prior professional experience in technology journalism
Must be living in and eligible to work in the United States
Should expect to travel two to three times per year for major event coverage
Must be comfortable with fully remote work

The full job description and official details can all be found at the listing on the Condé Careers site. If this sounds like the job for you, please apply!

Ars is seeking a seasoned senior reporter for all things Google Read More »

The 2024 Volkswagen ID.4 Pro gets a new rear motor, way more efficiency

car review, Cars, Volkswagen ID.4, VW ID.4 / Shannon Garcia / July 25, 2024

not selling well though —

40 percent more power, 30 percent more torque, and a range boost to boot.

Jonathan M. Gitlin – Jul 25, 2024 5: 41 pm UTC

A silver VW ID.4 next to some graffiti in an alley — Enlarge / The VW ID.4 has a new drive motor and infotainment system for model-year 2024. It’s not the sportiest EV you can buy, but it remains one of our favorites to drive.

Jonathan Gitlin

Volkswagen didn’t wait the traditional four model years before giving its ID.4 electric crossover something of a spiff-up. The tweaks to the model-year 2024 ID.4 are mostly under the skin or inside the cabin—like the recent refresh of the Polestar 2, this update was more about making the ID.4 an easier EV to live with, with more range and more power.

Volkswagen was one of the first automakers to react to Tesla finally making the electric vehicle viable. After the company-wide bet on diesel went up in a cloud of nitrogen oxides and black smoke, VW threw itself headlong into electrification as a way to meet ever-stricter carbon emissions regulations. Already an industry pioneer for the use of highly flexible vehicle architectures that let it build vehicles in a wide range of sizes and shapes with a common set of components and tools, it applied that approach to a line of electric vehicles, all branded under the Intelligent Design, or ID, name.

VW is a global automaker, but automobile tastes are often not global. For Europe, VW designed the ID.3, an electric hatchback that Americans who want forbidden fruit keep asking for, but which generated less than enthusiastic reviews from the people who actually got to buy them. Other models are optimized for China. But for America, with its adoration of the SUV and crossover, VW designed the ID.4.

The ID.4 was designed with America's love of crossovers in mind. — Enlarge / The ID.4 was designed with America’s love of crossovers in mind.

Jonathan Gitlin

Unveiled in the depths of the pandemic, we got our first (if short) drive in a prototype ID.4 in October 2020. Four months later, it was time to try the production version, an EV we proclaimed “a solid effort.” A few months later, we tried out the all-wheel drive ID.4 and checked out VW’s factory in Chattanooga, Tennessee, which has been churning out locally made ID.4s since mid-2022.

What’s new?

For model-year 2024, the $44,875 ID.4 Pro keeps its 82 kWh battery pack, but its EPA range jumps to 291 miles (468 km), up from 275 miles (443 km). The reason will probably seem counterintuitive if all you’re used to is gasoline cars—a new, more powerful drive unit that generates 282 hp (210 kW) and 402 lb-ft (545 Nm). That’s a 40 percent increase in power and a 30 percent increase in torque compared to the rear-wheel drive ID.4 Pro we tested in the past.

With internal-combustion engine vehicles, turning up the wick on the power and torque usually means your range plummets. Not so with an EV. The new motor has an improved stator and a new water- and oil-cooling system, both of which mean it can cope better with higher thermal loads—VW says this is “an elementary contributing factor” to the improved efficiency. The one-speed transmission has had its components optimized to reduce friction, and there’s a new inverter with all-new software.

A very tight turning circle means this is a great EV for American cities.

Jonathan Gitlin
There’s plenty of room here for your kids’ sports equipment, the monthly Costco run, or a vacation’s worth of luggage.

Jonathan Gitlin

The battery can also accept a higher rate of power during charging and regenerative braking, now 175 kW, up from 125 kW. That means a 10–80 percent fast charge should take 30 min. While we weren’t able to deplete the battery quite enough to test that, a charge from 35–80 percent state of charge took just 22 minutes at a peak of 155 kW, and just under 31 minutes was sufficient to reach a 92 percent SoC from that starting point.

On the road, and despite its mainstream design, the ID.4 remains a pretty good EV to drive. It has a very tight turning circle (31.5 feet/9/6 m), which is helpful in the city, and on a winding back road it is far better-mannered than a family crossover should be. A shared vehicle dynamics control system with the latest Golf GTI no doubt helps here.

I prefer Comfort mode over Sport; the latter makes the steering heavier but with no more feedback and makes the lift-off regen braking more aggressive. Power delivery is very smooth despite the bump in output.

In any of the three modes (which includes Eco as well as Comfort and Sport) the ride is a little bouncy—US market ID.4s do without adaptive dampers, so it doesn’t change when you switch. And there was a fair bit of road noise from the tires at highway speeds.

I was surprised that, in relatively mild weather, I was able to achieve an average of 4.1 miles/kWh (15.2 kWh/100 km). As the weather got hot and AC was a necessity, this dropped to 3.5 miles/kWh (17.8 kWh/100 km), which is still an improvement on the First Edition we tested in 2022.

The 2024 Volkswagen ID.4 Pro gets a new rear motor, way more efficiency Read More »

AMD delays Ryzen 9000 launch to August “out of an abundance of caution”

AMD, ryzen 9000, Tech, zen 5 / Shannon Garcia / July 24, 2024

ryzen is slippen —

More rigorous testing and screening have reportedly corrected the problem.

Andrew Cunningham – Jul 24, 2024 8: 09 pm UTC

AMD had planned to launch its first round of Ryzen 9000-series desktop processors by the end of July, but those plans have changed thanks to a very non-specific problem found with the first batch of processors that AMD sent out to its partners. The six- and eight-core Ryzen 9600X and 9700X are now slated to launch on August 8, and the 12- and 16-core Ryzen 9900X and 9950X will launch on August 15.

AMD’s full statement is below:

We appreciate the excitement around Ryzen 9000 series processors. During final checks, we found the initial production units that were shipped to our channel partners did not meet our full quality expectations. Out of an abundance of caution and to maintain the highest quality experiences for every Ryzen user, we are working with our channel partners to replace the initial production units with fresh units. As a result, there will be a short delay in retail availability. The Ryzen 7 9700X and Ryzen 5 9600X processors will now go on sale on August 8th and the Ryzen 9 9950X and Ryzen 9 9900X processors will go on-sale on August 15th. We pride ourselves in providing a high-quality experience for every Ryzen user, and we look forward to our fans having a great experience with the new Ryzen 9000 series.

When asked for details about the specific problem and what the fix was, AMD Public Relations Manager Matthew Hurwitz told Ars that AMD had implemented additional screening for the Ryzen 9000 CPUs but couldn’t share specifics about what AMD is screening for.

It doesn’t seem as though any changes are being made to the silicon or the manufacturing process itself, and Hurwitz told us that the first batch of processors would be sent back out to channel partners once they had been recalled and re-screened.

The Ryzen 9000-series CPUs are the direct follow-up to the Ryzen 7000 series from late 2022, and the second generation of chips to use the AM5 processor socket—if you don’t count the Ryzen 7000X3D CPUs, which are Ryzen 7000 with more L3 cache, or the Ryzen 8000G chips, which are Zen 4-based laptop processors repackaged for desktops. Ryzen 9000 chips should drop into existing AM5 motherboards after a BIOS update, though AMD is also releasing a mildly improved lineup of chipsets to power new boards.

AMD has prioritized power efficiency for the Ryzen 9000 chips but is still promising low-double-digit performance improvements in both single- and multi-core workloads.

AMD delays Ryzen 9000 launch to August “out of an abundance of caution” Read More »

Elon Musk claims he is training “the world’s most powerful AI by every metric”

AI, Biz & IT, chatgpt, chatgtp, elon musk, GPUs, grok, H100, large language models, Memphis, Memphis supercluster, NVIDIA, tennessee, Tesla, TVA, WREG, xAI / Shannon Garcia / July 24, 2024

the biggest, most powerful —

One snag: xAI might not have the electrical power contracts to do it.

Benj Edwards – Jul 24, 2024 4: 19 pm UTC

Enlarge / Elon Musk, chief executive officer of Tesla Inc., during a fireside discussion on artificial intelligence risks with Rishi Sunak, UK prime minister, in London, UK, on Thursday, Nov. 2, 2023.

On Monday, Elon Musk announced the start of training for what he calls “the world’s most powerful AI training cluster” at xAI’s new supercomputer facility in Memphis, Tennessee. The billionaire entrepreneur and CEO of multiple tech companies took to X (formerly Twitter) to share that the so-called “Memphis Supercluster” began operations at approximately 4: 20 am local time that day.

Musk’s xAI team, in collaboration with X and Nvidia, launched the supercomputer cluster featuring 100,000 liquid-cooled H100 GPUs on a single RDMA fabric. This setup, according to Musk, gives xAI “a significant advantage in training the world’s most powerful AI by every metric by December this year.”

Given issues with xAI’s Grok chatbot throughout the year, skeptics would be justified in questioning whether those claims will match reality, especially given Musk’s tendency for grandiose, off-the-cuff remarks on the social media platform he runs.

Power issues

According to a report by News Channel 3 WREG Memphis, the startup of the massive AI training facility marks a milestone for the city. WREG reports that xAI’s investment represents the largest capital investment by a new company in Memphis’s history. However, the project has raised questions among local residents and officials about its impact on the area’s power grid and infrastructure.

WREG reports that Doug McGowen, president of Memphis Light, Gas and Water (MLGW), previously stated that xAI could consume up to 150 megawatts of power at peak times. This substantial power requirement has prompted discussions with the Tennessee Valley Authority (TVA) regarding the project’s electricity demands and connection to the power system.

The TVA told the local news station, “TVA does not have a contract in place with xAI. We are working with xAI and our partners at MLGW on the details of the proposal and electricity demand needs.”

The local news outlet confirms that MLGW has stated that xAI moved into an existing building with already existing utility services, but the full extent of the company’s power usage and its potential effects on local utilities remain unclear. To address community concerns, WREG reports that MLGW plans to host public forums in the coming days to provide more information about the project and its implications for the city.

For now, Tom’s Hardware reports that Musk is side-stepping power issues by installing a fleet of 14 VoltaGrid natural gas generators that provide supplementary power to the Memphis computer cluster while his company works out an agreement with the local power utility.

As training at the Memphis Supercluster gets underway, all eyes are on xAI and Musk’s ambitious goal of developing the world’s most powerful AI by the end of the year (by which metric, we are uncertain), given the competitive landscape in AI at the moment between OpenAI/Microsoft, Amazon, Apple, Anthropic, and Google. If such an AI model emerges from xAI, we’ll be ready to write about it.

This article was updated on July 24, 2024 at 1: 11 pm to mention Musk installing natural gas generators onsite in Memphis.

Elon Musk claims he is training “the world’s most powerful AI by every metric” Read More »

Woman who went on the lam with untreated TB is now cured

antibiotics, infectious diseases, Science, Tacoma-Pierce County Health Department, tb, Tuberculosis, Washington / Shannon Garcia / July 24, 2024

happy ending —

The woman realized how serious her infection was once she was in custody.

Beth Mole – Jul 24, 2024 3: 01 pm UTC

Scanning electron micrograph of <em>Mycobacterium tuberculosis</em> bacteria, which cause TB.” src=”https://cdn.arstechnica.net/wp-content/uploads/2015/03/5149398678_97948614ea_o-514×640.jpg”></img><figcaption>
<p>Scanning electron micrograph of <em>Mycobacterium tuberculosis</em> bacteria, which cause TB.</p>
</figcaption></figure>
<p>“She’s cured!”</p>
<p>Health officials in Washington state are celebrating the clean bill of health for one particularly notable resident: the woman <a href=

Woman who went on the lam with untreated TB is now cured Read More »

Google halts its 4-plus-year plan to turn off tracking cookies by default in Chrome

Apple, chrome, Firefox, Floc, Google, google chrome, mozilla, privacy sandbox, Tech, third party cookies, tracking cookies / Shannon Garcia / July 23, 2024

Filling, but not nutritious —

A brief history of Google’s ideas, proposals, and APIs for cookie replacements.

Kevin Purdy – Jul 23, 2024 5: 22 pm UTC

A woman in a white knit sweater, holding a Linzer cookie (with jam inside a heart cut-out) in her crossed palms. — Enlarge / Google, like most of us, has a hard time letting go of cookies. Most of us just haven’t created a complex set of APIs and brokered deals across regulation and industry to hold onto the essential essence of cookies.

Getty Images

Google has an announcement today: It’s not going to do something it has thought about, and tinkered with, for quite some time.

Most people who just use the Chrome browser, rather than develop for it or try to serve ads to it, are not going to know what “A new path for Privacy Sandbox on the web” could possibly mean. The very short version is that Google had a “path,” first announced in January 2020, to turn off third-party (i.e., tracking) cookies in the most-used browser on Earth, bringing it in line with Safari, Firefox, and many other browsers. Google has proposed several alternatives to the cookies that follow you from page to page, constantly pitching you on that space heater you looked at three days ago. Each of these alternatives has met varying amounts of resistance from privacy and open web advocates, trade regulators, and the advertising industry.

So rather than turn off third-party cookies by default and implement new solutions inside the Privacy Sandbox, Chrome will “introduce a new experience” that lets users choose their tracking preferences when they update or first use Chrome. Google will also keep working on its Privacy Sandbox APIs but in a way that recognizes the “impact on publishers, advertisers, and everyone involved in online advertising.” Google also did not fail to mention it was “discussing this new path with regulators.”

Why today? What does it really mean? Let’s journey through more than four and a half years of Google’s moves to replace third-party cookies, without deeply endangering its standing as the world’s largest advertising provider.

2017–2022: FLoC or “What if machines tracked you, not cookies?”

Google’s big moves toward a standstill likely started at Apple headquarters. Its operating system updates in the fall of 2017 implemented a 24-hour time limit on ad-targeting cookies in Safari, the default browser on Macs and iOS devices. A “Coalition of Major Advertising Trade Associations” issued a sternly worded letter opposing this change, stating it would “drive a wedge between brands and their customers” and make advertising “more generic and less timely and useful.”

By the summer of 2019, Firefox was ready to simply block tracking cookies by default. Google, which makes the vast majority of its money through online advertising, made a different, broader argument against dropping third-party cookies. To paraphrase: Trackers will track, and if we don’t give them a proper way to do it, they’ll do it the dirty way by fingerprinting browsers based on version numbers, fonts, screen size, and other identifiers. Google said it had some machine learning that could figure out when it was good to share your browsing habits. For example:

New technologies like Federated Learning show that it’s possible for your browser to avoid revealing that you are a member of a group that likes Beyoncé and sweater vests until it can be sure that group contains thousands of other people.

In January 2020, Google shifted its argument from “along with” to “instead of” third-party cookies. Chrome Engineering Director Justin Schuh wrote, “Building a more private Web: A path towards making third party cookies obsolete,” suggesting that broad support for Chrome’s privacy sandbox tools would allow for dropping third-party cookies entirely. Privacy advocate Ben Adida described the move as “delivering teeth” and “a big deal.” Feedback from the W3C and other parties, Schuh wrote at that time, “gives us confidence that solutions in this space can work.”

Google's explanatory graphic for FLoC, or Federated Learning of Cohorts. — Google’s explanatory graphic for FLoC, or Federated Learning of Cohorts.

Google

As Google developed its replacement for third-party cookies, the path grew trickier and the space more perilous. The Electronic Frontier Foundation described Google’s FLoC, or the “Federated Learning of Cohorts” that would let Chrome machine-learn your profile for sites and ads, as “A Terrible Idea.” The EFF was joined by Mozilla, Apple, WordPress, DuckDuckGo, and lots of browsers based on Chrome’s core Chromium code in being either opposed or non-committal to FLoC. Google pushed back testing FLOC until late 2022 and third-party cookie removal (and thereby FLoC implementation) until mid-2023.

By early 2022, FLoC didn’t have a path forward. Google pivoted to a Topics API, which would give users a bit more control over which topics (“Rock Music,” “Auto & Vehicles”) would be transmitted to potential advertisers. It would certainly improve over third-party cookies, which are largely inscrutable in naming and offer the user only one privacy policy: block them, or delete them all and lose lots of logins.

Google halts its 4-plus-year plan to turn off tracking cookies by default in Chrome Read More »

Author name: Shannon Garcia