Ars Technica
Elon Musk’s Neuralink reports trouble with first human brain chip
Enlarge / Elon Musk, in Paris, France, on Friday, June 16, 2023. Musk predicted his Neuralink Corp. would carry out its first brain implant later that year. The first implantation took place in January 2024. (credit: Getty | Nathan Laine)
The first invasive brain chip that Neuralink embedded into a human brain has malfunctioned, with neuron-surveilling threads appearing to have become dislodged from the participant's brain, the company revealed in a blog post Wednesday.
It's unclear what caused the threads to become "retracted" from the brain, how many have retracted, or if the displaced threads pose a safety risk. Neuralink, the brain-computer interface startup run by controversial billionaire Elon Musk, did not immediately respond to a request for comment from Ars. The company said in its blog post that the problem began in late February, but it has since been able to compensate for the lost data to some extent by modifying its algorithm.
Neuralink touts that its invasive implant includes 64 flexible threads carrying a total of 1,024 electrodes that can detect neuronal activity. Those flexible threads—described as thinner than a human hair—are inserted individually into the brain by the company's proprietary surgical robot. The goal is for the threads to be placed near neurons of interest so that signals detected by the electrodes can be recorded and decoded into intended actions, such as moving a cursor on a computer screen.
Max, Disney+, Hulu ad-free bundle coming amid high streaming cancellation rates
Enlarge / Streaming services are assembling... into a bundled package. (credit: Marvel Entertainment/YouTube)
Sometime this summer, US customers will be able to buy a subscription to Max, Disney+, and Hulu together for a discounted price. The Wednesday announcement from respective owners Warner Bros. Discovery (WBD) and Disney comes as the streaming industry combats a competitive subscription marketplace burdened by constant cancellations.
WBD and Disney didn't provide a specific release date for the package but said that people will be able to buy it from "any of the three streaming platform’s websites offered as both an ad-supported and ad-free plan.”
The companies didn't confirm a price, but the bundle should be cheaper than all three services combined, which would start at $47.97 per month for no ads and $25.97/month with ads.
A crushing backlash to Apple’s new iPad ad
Enlarge / A screenshot of the Apple iPad ad. (credit: Apple via YouTube)
An advert by Apple for its new iPad tablet showing musical instruments, artistic tools, and games being crushed by a giant hydraulic press has been attacked for cultural insensitivity in an online backlash.
The one-minute video was launched by Apple chief executive Tim Cook to support its new range of iPads, the first time that the US tech giant has overhauled the range for two years as it seeks to reverse faltering sales.
The campaign—soundtracked by Sonny and Cher’s 1971 hit All I Ever Need Is You—is designed to show how much Apple has been able to squeeze into the thinner tablet. The ad was produced in-house by Apple’s creative team, according to trade press reports.
NASA confirms “independent review” of Orion heat shield issue
Enlarge / The Orion spacecraft after splashdown in the Pacific Ocean at the end of the Artemis I mission. (credit: NASA)
NASA has asked a panel of outside experts to review the agency's investigation into the unexpected loss of material from the heat shield of the Orion spacecraft on a test flight in 2022.
Chunks of charred material cracked and chipped away from Orion's heat shield during reentry at the end of the 25-day unpiloted Artemis I mission in December 2022. Engineers inspecting the capsule after the flight found more than 100 locations where the stresses of reentry stripped away pieces of the heat shield as temperatures built up to 5,000° Fahrenheit.
This was the most significant discovery on the Artemis I, an unpiloted test flight that took the Orion capsule around the Moon for the first time. The next mission in NASA's Artemis program, Artemis II, is scheduled for launch late next year on a test flight to send four astronauts around the far side of the Moon.
Professor sues Meta to allow release of feed-killing tool for Facebook
Enlarge (credit: themotioncloud/Getty Images)
Ethan Zuckerman wants to release a tool that would allow Facebook users to control what appears in their newsfeeds. His privacy-friendly browser extension, Unfollow Everything 2.0, is designed to essentially give users a switch to turn the newsfeed on and off whenever they want, providing a way to eliminate or curate the feed.
Ethan Zuckerman, a professor at University of Massachusetts Amherst, is suing Meta to release a tool allowing users to "unfollow everything" on Facebook.
The tool is nearly ready to be released, Zuckerman told Ars, but the University of Massachusetts Amherst associate professor is afraid that Facebook owner Meta might threaten legal action if he goes ahead. And his fears appear well-founded. In 2021, Meta sent a cease-and-desist letter to the creator of the original Unfollow Everything, Louis Barclay, leading that developer to shut down his tool after thousands of Facebook users had eagerly downloaded it.
Zuckerman is suing Meta, asking a US district court in California to invalidate Meta's past arguments against developers like Barclay and rule that Meta would have no grounds to sue if he released his tool.
Logic Pro gets some serious AI—and a version bump—for Mac and iPad
Enlarge / The new Chord Track feature. (credit: Apple)
If you watched yesterday's iPad-a-palooza event from Apple, then you probably saw the segment about cool new features coming to the iPad version of Logic Pro, Apple's professional audio recording software. But what the event did not make clear was that all the same features are coming to the Mac version of Logic Pro—and both the Mac and iPad versions will get newly numbered. After many years, the Mac version of Logic Pro will upgrade from X (ten) to 11, while the much more recent iPad version increments to 2.
Both versions will be released on May 13, and both are free upgrades for existing users. (Sort of—iPad users have to pay a subscription fee to access Logic Pro, but if you already pay, you'll get the upgrade. This led many people to speculate online that Apple would move the Mac version of Logic to a similar subscription model; thankfully, that is not the case. Yet.)
Both versions will gain an identical set of new features, which were touched on briefly in Apple's event video. But thanks to a lengthy press release that Apple posted after the event, along with updates to Apple's main Logic page, we now have a better sense of what these features are, what systems they require, and just how much Apple has gone all-in on AI. Also, we get some pictures.
Critical vulnerabilities in BIG-IP appliances leave big networks open to intrusion
Enlarge (credit: Getty Images)
Researchers on Wednesday reported critical vulnerabilities in a widely used networking appliance that leaves some of the world’s biggest networks open to intrusion.
The vulnerabilities reside in BIG-IP Next Central Manager, a component in the latest generation of the BIG-IP line of appliances, which organizations use to manage traffic going into and out of their networks. Seattle-based F5, which sells the product, says its gear is used in 48 of the top 50 corporations as tracked by Fortune. F5 describes the Next Central Manager as a “single, centralized point of control” for managing entire fleets of BIG-IP appliances.
As devices performing load balancing, DDoS mitigation, and inspection and encryption of data entering and exiting large networks, BIG-IP gear sits at their perimeter and acts as a major pipeline to some of the most security-critical resources housed inside. Those characteristics have made BIG-IP appliances ideal for hacking. In 2021 and 2022, hackers actively compromised BIG-IP appliances by exploiting vulnerabilities carrying severity ratings of 9.8 out of 10.
Dell responds to return-to-office resistance with VPN, badge tracking
After reversing its positioning on remote work, Dell is reportedly implementing new tracking techniques on May 13 to ensure its workers are following the company's return-to-office (RTO) policy, The Register reported today, citing anonymous sources.
Dell has allowed people to work remotely for over 10 years. But in February, it issued an RTO mandate, and come May 13, most workers will be classified as either totally remote or hybrid. Starting this month, hybrid workers have to go into a Dell office at least 39 days per quarter. Fully remote workers, meanwhile, are ineligible for promotion, Business Insider reported in March.
Now The Register reports that Dell will track employees' badge swipes and VPN connections to confirm that workers are in the office for a significant amount of time.
40 years later, Kontrabant 2 for ZX Spectrum is rebroadcast on FM in Slovenia
Enlarge / In 1984, the year 2000 was so promising, students made entire games promising to take you there. (credit: Radio Student)
Software is almost impossibly easy to download, distribute, and access compared to 40 years ago. Everything is bigger, faster, and more flexible, but there's a certain charm to the ways of diskettes and cassettes that is hard to recapture. That doesn't mean we can't try.
By the time you read this, it's likely that Kontrabant 2 will have already hit the airwaves on Radio Študent in Slovenia. At 9:30 pm Slovenia time (UTC+2 in Daylight Savings Time), if you are tuned to 89.3 FM, hitting record on a cassette tape will capture a buzzing sound that will run until just over 50KB have been transmitted. If all went well, you can load the tape into your working ZX Spectrum or bring it to the Computer History Museum in Slovenia and use theirs to try it out.
Robot dogs armed with AI-targeting rifles undergo US Marines Special Ops evaluation
Enlarge / A still image of a robotic quadruped armed with a remote weapons system, captured from a video provided by Onyx Industries. (credit: Onyx Industries)
The United States Marine Forces Special Operations Command (MARSOC) is currently evaluating a new generation of robotic "dogs" developed by Ghost Robotics, with the potential to be equipped with gun systems from defense tech company Onyx Industries, reports The War Zone.
While MARSOC is testing Ghost Robotics' quadrupedal unmanned ground vehicles (called "Q-UGVs" for short) for various applications, including reconnaissance and surveillance, it's the possibility of arming them with weapons for remote engagement that may draw the most attention. But it's not unprecedented: The US Marine Corps has also tested robotic dogs armed with rocket launchers in the past.
MARSOC is currently in possession of two armed Q-UGVs undergoing testing, as confirmed by Onyx Industries staff, and their gun systems are based on Onyx's SENTRY remote weapon system (RWS), which features an AI-enabled digital imaging system and can automatically detect and track people, drones, or vehicles, reporting potential targets to a remote human operator that could be located anywhere in the world. The system maintains a human-in-the-loop control for fire decisions, and it cannot decide to fire autonomously.
DeepMind adds a diffusion engine to latest protein-folding software
Enlarge / Prediction of the structure of a coronavirus Spike protein from a virus that causes the common cold. (credit: Google DeepMind)
Most of the activities that go on inside cells—the activities that keep us living, breathing, thinking animals—are handled by proteins. They allow cells to communicate with each other, run a cell's basic metabolism, and help convert the information stored in DNA into even more proteins. And all of that depends on the ability of the protein's string of amino acids to fold up into a complicated yet specific three-dimensional shape that enables it to function.
Up until this decade, understanding that 3D shape meant purifying the protein and subjecting it to a time- and labor-intensive process to determine its structure. But that changed with the work of DeepMind, one of Google's AI divisions, which released Alpha Fold in 2021, and a similar academic effort shortly afterward. The software wasn't perfect; it struggled with larger proteins and didn't offer high-confidence solutions for every protein. But many of its predictions turned out to be remarkably accurate.
Even so, these structures only told half of the story. To function, almost every protein has to interact with something else—other proteins, DNA, chemicals, membranes, and more. And, while the initial version of AlphaFold could handle some protein-protein interactions, the rest remained black boxes. Today, DeepMind is announcing the availability of version 3 of AlphaFold, which has seen parts of its underlying engine either heavily modified or replaced entirely. Thanks to these changes, the software now handles a variety of additional protein interactions and modifications.
Intel’s and Qualcomm’s Huawei export licenses get revoked
Enlarge / Huawei's Intel-powered Matebook X Pro has drawn criticism from US China hawks. (credit: Huawei)
The US crackdown on exports to Huawei now includes even stronger restrictions than the company has already faced. The Financial Times reports that Intel and Qualcomm have had their Huawei export licenses revoked, so Huawei will no longer be able to buy chips from either company.
The export ban has been around since 2020 and means that any company wishing to ship parts to Huawei must get approval from the government on a case-by-case basis. Sometimes these come with restrictions, like Qualcomm's license, which allowed it to ship smartphone chips to Huawei, but not "5G" chips. That led to Qualcomm creating special 4G-only versions of its 5G chips for Huawei, and the company ended up with 4G-only Snapdragon 888 phones in 2021.
Since then, Huawei has been working on its own Arm chips from its chip design division, HiSilicon. In April, the Huawei Pura 70 smartphone launched with an in-house HiSilicon Kirin 9010 SoC made at SMIC, a Chinese chip fab that is also facing export restrictions. With what is probably still a 7 nm manufacturing process, it's more of a 2020 chip than a 2024 chip, but that's still fast enough for many use cases.
Bike brands start to adopt C-V2X to warn cyclists about cars
Enlarge / A C-V2X bicycle can alert a C-V2X-enabled car to its presence—note the large yellow icon on the screenshot of the Audi's digital instrument display. (credit: Audi)
There's a fundamental flaw in current car safety tech: It's limited to line of sight. Or, perhaps, line of "sensing" is more accurate, because the way cameras and lidar work is to inspect the perimeter of a vehicle and use predictive algorithms to understand the motion of an object in relation to the motion of the vehicle itself. Which is good, because as carmakers have added elements such as pedestrian and cyclist detection, they're trying to prevent drivers from hurting the most vulnerable road users. And unfortunately this is necessary, because even though 2023 saw a slight reduction in drivers striking cyclists and pedestrians, according to the most recent data from the Governor's Highway Safety Association, since 2019 pedestrian fatalities are still up 14 percent—and cyclist deaths are up 50 percent since 2010.
That doesn't mean lidar and cameras have "failed," but because they rely on what the sensors can pick up, they cannot necessarily ID hazards (and alert drivers) as quickly as we need them to, particularly if that's a cyclist in your lane 300 feet down the road, just over the next rise. Yes, current sensing works well now with figuring out the pace of a traffic jam, and automatic emergency braking can step in to stop your car if you fail to. But for non-automotive obstacles, they're still limited.
For that, we need better tech, which is emerging and is called Connected Vehicle to Everything (C-V2X). The idea isn't that complicated. Boiled down, it's a chipset that operates on a portion of the cellular bandwidth, and vehicles with this tech embedded (say in an e-bike or car) monitor anything with a C-V2X chip as well as broadcast their own location at a pulse of 10 times a second. This precision location system would then warn a driver of a cyclist on the road ahead, even beyond line of sight, and in an emergency—possibly because a cyclist was right in a car's path—could prevent a collision.
M4 iPad Pro CPU cores and RAM amount are tied to storage capacity
Enlarge / The new M4 iPad Pro. (credit: Apple)
When Apple announced the Apple M4 chip during its iPad Pro event yesterday, it mentioned that the chip came with "up to" four high-performance CPU cores.
Those short, easily missable words always mean that there's a lower-end version of the chip coming that doesn't include that many CPU cores, and the tech specs page for the new iPad Pro has the full details: iPad Pros with 256GB or 512GB of storage use a version of the M4 with three high-performance CPU cores and six smaller efficiency cores. Only the models with 1TB and 2TB of storage have an M4 with all four high-performance CPU cores enabled.
The 256GB and 512GB models also ship with 8GB of RAM, where the 1TB and 2TB models come with 16GB of memory installed. Though these changes are clearly spelled out on the Tech Specs page, the actual configuration page for the iPad Pros in Apple's online store doesn't give any indication that upgrading storage also upgrades your CPU and RAM.
FCC explicitly prohibits fast lanes, closing possible net neutrality loophole
Enlarge (credit: Getty Images | nadla)
The Federal Communications Commission clarified its net neutrality rules to prohibit more kinds of fast lanes.
While the FCC voted to restore net neutrality rules on April 25, it didn't release the final text of the order until yesterday. The final text has some changes compared to the draft version released a few weeks before the vote.
Both the draft and final rules ban paid prioritization, or fast lanes that application providers have to pay Internet service providers for. But some net neutrality proponents raised concerns about the draft text because it would have let ISPs speed up certain types of applications as long as the application providers don't have to pay for special treatment.
Tesla is under a federal wire fraud probe for misleading investors
Enlarge (credit: Aurich Lawson | Tesla | Airplane!)
There's more bad news for Tesla. On Monday, we learned that CEO Elon Musk is continuing to slash his way through the company payroll as Tesla went through a fourth round of layoffs in four weeks. Yesterday, we discovered exactly what questions the National Highway Traffic Safety Administration wants answered about the safety of Tesla's Autopilot driver assist. And today, it emerged that the US Department of Justice is investigating whether or not Tesla committed securities or wire fraud by making misleading statements about Autopilot and its so-called "Full Self-Driving" (FSD) option.
Reuters reported that three people familiar with the matter told it about the investigation. One of the sources also told Reuters that the Securities and Exchange Commission is also investigating Tesla's claims about its driver assists.
Not the first timeThis isn't the first time Tesla has been accused of securities fraud. In 2018, Musk agreed to a settlement with the SEC over his infamous "funding secured" tweet that sent the company's share price skyrocketing despite the fact that there was never actually a possibility that he would take the company private. As a result, Musk was required to step down as chairman, and both Musk and Tesla were ordered to pay $20 million in penalties, to be distributed to investors who lost money after being misled by Musk.
Court rules against Activision Blizzard in $23.4M patent dispute
Enlarge / Acceleration Bay says World of Warcraft's networking code infringes on a patent originally filed by Boeing. (credit: Activision Blizzard)
A jury has found Activision Blizzard liable for $23.4 million in damages in a patent infringement lawsuit first brought to court in 2015.
The case centers on patents first filed by Boeing in 2000, one that describes a "distributed game environment" across a host and multiple computers and another that describes a simple method for disconnecting from such a network. Those patents were acquired in 2015 by Acceleration Bay, which accused Activision Blizzard of using infringing technology to develop World of Warcraft and at least two Call of Duty titles.
Those accusations succeeded in court earlier this week, as a jury found a "preponderance of evidence" that the patents were infringed. The decision came following a one-week trial in which Activision Blizzard argued that its networking technology works differently from what is described in the patents, as reported by Reuters.
No one has seen the data behind Tyson’s “climate friendly beef” claim
Enlarge / The Environmental Working Group published a new analysis on Wednesday outlining its efforts to push the USDA for more transparency, including asking for specific rationale in allowing brands to label beef as “climate friendly.” (credit: Carolyn Van Houten/Washington Post via Getty)
About five miles south of Broken Bow, in the heart of central Nebraska, thousands of cattle stand in feedlots at Adams Land & Cattle Co., a supplier of beef to the meat giant Tyson Foods.
From the air, the feedlots look dusty brown and packed with cows—not a vision of happy animals grazing on open pastureland, enriching the soil with carbon. But when the animals are slaughtered, processed, and sent onward to consumers, labels on the final product can claim that they were raised in a “climate friendly” way.
In late 2022, Tyson—one of the country’s “big four” meat packers—applied to the US Department of Agriculture (USDA), seeking a “climate friendly” label for its Brazen Beef brand. The production of Brazen Beef, the label claims, achieves a “10 percent greenhouse gas reduction.” Soon after, the USDA approved the label.
Prime Video subs will soon see ads for Amazon products when they hit pause
Enlarge / A scene from the Prime Video original series Fallout. (credit: Prime Video/YouTube)
Amazon Prime Video subscribers will see new types of advertisements this broadcast year. Amazon announced today that it's adding new ad formats to its video streaming service, hoping to encourage people to interact with the ads and shop on Amazon.
In January, Prime Video streams included commercials unless subscribers paid $3 extra per month. That has meant that watching stuff on Prime Video ad-free costs $12 per month or, if you're also a Prime subscriber, $18 per month.
New types of Prime Video adsAmazon has heightened focus on streaming ads this year. Those who opted for Prime Video with commercials will soon see shoppable carousel ads, interactive pause ads, and interactive brand trivia ads, as Amazon calls them. Amazon said that advertisers could buy these new displays to be shown "across the vast majority of content on Prime Video, wherever it’s streamed." All the new ad formats allow a viewer to place advertised products in their Amazon cart.
OpenAI’s flawed plan to flag deepfakes ahead of 2024 elections
Enlarge (credit: Boris Zhitkov | Moment)
As the US moves toward criminalizing deepfakes—deceptive AI-generated audio, images, and videos that are increasingly hard to discern from authentic content online—tech companies have rushed to roll out tools to help everyone better detect AI content.
But efforts so far have been imperfect, and experts fear that social media platforms may not be ready to handle the ensuing AI chaos during major global elections in 2024—despite tech giants committing to making tools specifically to combat AI-fueled election disinformation. The best AI detection remains observant humans, who, by paying close attention to deepfakes, can pick up on flaws like AI-generated people with extra fingers or AI voices that speak without pausing for a breath.
Among the splashiest tools announced this week, OpenAI shared details today about a new AI image detection classifier that it claims can detect about 98 percent of AI outputs from its own sophisticated image generator, DALL-E 3. It also "currently flags approximately 5 to 10 percent of images generated by other AI models," OpenAI's blog said.