Airbus, Leonardo, and Thales are merging their space divisions into a new France-based company that aims to create a "leading European player in space." The joint venture, expected to launch operations by 2027 pending regulatory approval, will pool R&D resources to accelerate satellite development and strengthen Europe's technological sovereignty in space. Engadget reports: The companies Airbus, Leonardo and Thales have finalized this deal. The new unnamed entity will be based in France and will employ around 25,000 people. Airbus will own 35 percent, while the other two companies will each own 32.5 percent. Executives are hoping this company will better serve Europe's need for "sovereignty" in space and help it create a rival to SpaceX's Starlink communications network. Increasing a presence in space is also seen as a good thing for security and defense.

This isn't just bluster. Thales and Airbus have long been rivals in the satellite market, but it looks like they are friends now. Leonardo is known for space systems and services. Combining all three could actually give SpaceX a run for its money, but we will have to wait and see. There are no planned site closures, as the companies say that each home country will keep its existing capabilities. This will be a standalone company, so think of it as an extremely well-financed startup. The first task for the upstart? Reporting indicates it'll be to find more efficient ways to develop and manufacture satellites.

SpaceX has deactivated over 2,500 Starlink terminals allegedly used by scam operations in Myanmar, where the service isn't licensed but was reportedly enabling large-scale cybercrime networks tied to human trafficking and fraud. Ars Technica reports: Lauren Dreyer, vice president of Starlink business operations, described the action in an X post last night after reports that Myanmar's military shut down a major scam operation: "SpaceX complies with local laws in all 150+ markets where Starlink is licensed to operate," Dreyer wrote. "SpaceX continually works to identify violations of our Acceptable Use Policy and applicable law... On the rare occasion we identify a violation, we take appropriate action, including working with law enforcement agencies around the world. In Myanmar, for example, SpaceX proactively identified and disabled over 2,500 Starlink Kits in the vicinity of suspected 'scam centers.'"

Starlink is not licensed to operate in Myanmar. While Dreyer didn't say how the terminals were disabled, it's known that Starlink can disable individual terminals based on their ID numbers or use geofencing to block areas from receiving signals. On Monday, Myanmar state media reported that "Myanmar's military has shut down a major online scam operation near the border with Thailand, detaining more than 2,000 people and seizing dozens of Starlink satellite Internet terminals," according to an Associated Press article. The army reportedly raided a cybercrime center known as KK Park as part of operations that began in early September. The operations reportedly targeted 260 unregistered buildings and resulted in seizure of 30 Starlink terminals and detention of 2,198 people.

"Maj. Gen. Zaw Min Tun, the spokesperson for the military government, charged in a statement Monday night that the top leaders of the Karen National Union, an armed ethnic organization opposed to army rule, were involved in the scam projects at KK Park," the AP wrote. The Karen National Union is "part of the larger armed resistance movement in Myanmar's civil war" and "deny any involvement in the scams."

Google is migrating all its internal workloads to run on both x86 and its custom Axion Arm chips, with major services like YouTube, Gmail, and BigQuery already running on both architectures. The Register reports: The search and ads giant documented its move in a preprint paper published last week, titled "Instruction Set Migration at Warehouse Scale," and in a Wednesday post that reveals YouTube, Gmail, and BigQuery already run on both x86 and its Axion Arm CPUs -- as do around 30,000 more applications. Both documents explain Google's migration process, which engineering fellow Parthasarathy Ranganathan and developer relations engineer Wolff Dobson said started with an assumption "that we would be spending time on architectural differences such as floating point drift, concurrency, intrinsics such as platform-specific operators, and performance." [...]

The post and paper detail work on 30,000 applications, a collection of code sufficiently large that Google pressed its existing automation tools into service -- and then built a new AI tool called "CogniPort" to do things its other tools could not. [...] Google found the agent succeeded about 30 percent of the time under certain conditions, and did best on test fixes, platform-specific conditionals, and data representation fixes. That's not an enormous success rate, but Google has at least another 70,000 packages to port.

The company's aim is to finish the job so its famed Borg cluster manager -- the basis of Kubernetes -- can allocate internal workloads in ways that efficiently utilize Arm servers. Doing so will likely save money, because Google claims its Axion-powered machines deliver up to 65 percent better price-performance than x86 instances, and can be 60 percent more energy-efficient. Those numbers, and the scale of Google's code migration project, suggest the web giant will need fewer x86 processors in years to come.

Investigators recovered the OceanGate Titan sub's underwater camera nearly intact, discovering a SanDisk SD card that survived the 2023 implosion and still contained 12 images and 9 videos. TechSpot reports: Scott Manley, the science communication YouTuber, gamer, astrophysicist, and programmer, posted about the latest find: a hardened SubC-branded Rayfin Mk2 Benthic Camera containing the undamaged SD card. The titanium and synthetic sapphire crystal camera is rated to withstand depths of up to 6,000 meters (19,685 feet) -- the Titan imploded at around 3,300 meters (10,827 feet). The casing is intact, though the lens is shattered and the PCBs are slightly damaged.

Incredibly the SD card inside the camera was undamaged. Tom's Hardware reports that it's almost certainly a SanDisk Extreme Pro 512GB, which costs around $62 on Amazon. The camera's SD card was found to be fully encrypted, divided into a small partition for operating system updates and a larger one for user data. Due to impact damage from the accident, several components of the system-on-module (SOM) board -- including connectors and the microcontroller -- were broken, complicating the data extraction process. [...] After determining the data wasn't encrypted beyond the file system level, they successfully accessed the SD card contents using the manufacturer's proprietary equipment and procedures.

U.S. Grandmaster and beloved chess commentator Daniel Naroditsky has tragically passed away at the age of 29. "The news has sent shockwaves around the chess community, which is grieving the loss of one of the most beloved and influential voices," reports Chess.com. From the report: The devastating news was first shared by Naroditsky's club, Charlotte Chess Center, on Monday, and confirmed by Chess.com with multiple sources: "It is with great sadness that we share the unexpected passing of Daniel Naroditsky. Daniel was a talented chess player, educator, and cherished member of the chess community. He was also a loving son, brother, and loyal friend. We ask for privacy for Daniel's family during this extremely difficult time. Let us honor Daniel by remembering his passion for chess and the inspiration he brought to us all."

Naroditsky, who was three weeks away from turning 30, has long been known as one of United States' most talented players. He achieved his grandmaster title at the age of 18 in 2013, and placed fifth among the highest-ranked juniors in 2015. His last FIDE-rating is 2619, which places him among the top 150 in the world, or the 17th highest-ranked in the United States. He has a peak rating of 2647 from 2017. He leaves a legacy that spans strong over-the-board competition and highly popular chess instruction and commentary on streaming platforms.

Test scores across OECD countries peaked around 2012 and have declined since. IQ scores in many developed countries appear to be falling after rising throughout the twentieth century. Nataliya Kosmyna at MIT's Media Lab began noticing changes around two years ago when strangers started emailing her to ask if using ChatGPT could alter their brains. She posted a study in June tracking brain activity in 54 students writing essays. Those using ChatGPT showed significantly less activity in networks tied to cognitive processing and attention compared to students who wrote without digital help or used only internet search engines. Almost none could recall what they had written immediately after submitting their work.

She received more than 4,000 emails afterward. Many came from teachers who reported students producing passable assignments without understanding the material. A British survey found that 92% of university students now use AI and roughly 20% have used it to write all or part of an assignment. Independent research has found that more screen time in schools correlates with worse results. Technology companies have designed products to be frictionless, removing the cognitive challenges brains need to learn. AI now allows users to outsource thinking itself.

A network of classified Starshield satellites built by SpaceX for the U.S. government is transmitting signals on radio frequencies reserved for Earth-to-space commands. According to NPR, it may violate international standards. From the report: Satellites associated with the Starshield satellite network appear to be transmitting to the Earth's surface on frequencies normally used for doing the exact opposite: sending commands from Earth to satellites in space. The use of those frequencies to "downlink" data runs counter to standards set by the International Telecommunication Union, a United Nations agency that seeks to coordinate the use of radio spectrum globally.

Starshield's unusual transmissions have the potential to interfere with other scientific and commercial satellites, warns Scott Tilley, an amateur satellite tracker in Canada who first spotted the signals. "Nearby satellites could receive radio-frequency interference and could perhaps not respond properly to commands -- or ignore commands -- from Earth," he told NPR.

Outside experts agree there's the potential for radio interference. "I think it is definitely happening," said Kevin Gifford, a computer science professor at the University of Colorado, Boulder who specializes in radio interference from spacecraft. But he said the issue of whether the interference is truly disruptive remains unresolved. [...] Tilley says he's detected signals from 170 of the Starshield satellites so far. All appear in the 2025-2110 MHz range, though the precise frequencies of the signals move around.

An anonymous reader quotes a report from The Guardian: I realize that, at this point, there are already far too many shows. Every channel, every streaming service is teeming with content demanding your attention, and there are simply too few hours in the day to watch them all. However, with that in mind, may I recommend a new show called Space Live? There's only one episode. The only potential downside is that the episode literally lasts for ever. Actually, that's inaccurate. Space Live isn't a show, it's a channel. It launched on Wednesday morning, tucked away on ITVX, and consists only of live footage of Earth broadcast from the International Space Station. It's beguiling to watch, especially for anyone who didn't realize that a person can be awestruck and bored simultaneously.

It's billed as a world first. ITV has partnered with British space media company Sen to use live 4K footage from its proprietary SpaceTV-1 video camera system, mounted on the International Space Station, giving us three camera views: one of the station's docking ports, a horizon view able to show sunrises and storms, and a camera pointing straight down as the ISS passes across the planet. A tracker in the corner of the screen shows the live location of the ISS, while a real-time AI information feed provides facts about our geography and weather systems.

Of course, if you wanted to be picky, you could argue it isn't exactly new. Nasa's YouTube channel has been streaming live footage from the ISS for years, and uniformly draws an audience of a few thousand. But Space Live is, if nothing else, slightly snazzier. The footage is certainly nicer: at 8.30am on Wednesday, Space Live showed gorgeous images of the sun's glare bouncing off the sea around the Bay of Biscay, while all Nasa could offer was a piece of cloth with the word "Flap" written on it. There's even a soundtrack, a constant, soothing kind of hold music that loops and loops without ever becoming fully annoying. It's an improvement, in other words. And, at least for the first orbit, it is absolutely spellbinding.

EA employees and the Communications Workers of America union have condemned the company's proposed $55 billion private acquisition -- backed by Saudi Arabia's Public Investment Fund and Jared Kushner's Affinity Partners, "claiming they were not represented in the negotiations and any jobs lost as a result would 'be a choice, not a necessity, made to pad investors' pockets," reports Eurogamer. From the report: Following the announcement, there's been plenty of speculation around the future of EA and its multiple owned studios, split between EA Sports and EA Entertainment. Now, members of the United Videogame Workers union and the CWA have issued a formal response alongside a petition for regulators to scrutinize the deal. "EA is not a struggling company," the statement reads. "With annual revenues reaching $7.5 billion and $1 billion in profit each year, EA is one of the largest video game developers and publishers in the world."

This success has been driven by company workers, the union stated. "Yet we, the very people who will be jeopardized as a result of this deal, were not represented at all when this buyout was negotiated or discussed." Citing the number of layoffs across the industry since 2022, workers fear for "the future of our studios that are arbitrarily deemed 'less profitable' but whose contributions to the video game industry define EA's reputation." "If jobs are lost or studios are closed due to this deal, that would be a choice, not a necessity, made to pad investors' pockets - not to strengthen the company," the statement reads.

"Every time private equity or billionaire investors take a studio private, workers lose visibility, transparency, and power," it continues. "Decisions that shape our jobs, our art, and our futures are made behind closed doors by executives who have never written a line of code, built worlds, or supported live services. We are calling on regulators and elected officials to scrutinize this deal and ensure that any path forward protects jobs, preserves creative freedom, and keeps decision-making accountable to the workers who make EA successful." As such, workers have launched a petition in a "fight to make video games better for workers and players -- not billionaires". The statement concludes: "The value of video games is in their workers. As a unified voice, we, the members of the industry-wide video game workers' union UVW-CWA, are standing together and refusing to let corporate greed decide the future of our industry."

An anonymous reader quotes a report from 404 Media: On Monday, a publicly-sourced archive of more than 10,000 national park signs and monument placards went public as part of a massive volunteer project to save historical and educational placards from around the country that risk removal by the Trump administration. Visitors to national parks and other public monuments at more than 300 sites across the U.S. took photos of signs and submitted them to the archive to be saved in case they're ever removed in the wake of the Trump administration's rewriting of park history. The full archive is available here, with submissions from July to the end of September. The signs people have captured include historical photos from Alcatraz, stories from the African American Civil War Memorial, photos and accounts from the Brown v. Board of Education National History Park, and hundreds more sites. "I'm so excited to share this collaborative photo collection with the public. As librarians, our goal is to preserve the knowledge and stories told in these signs. We want to put the signs back in the people's hands," Jenny McBurney, Government Publications Librarian at the University of Minnesota and one of the co-founders of the Save Our Signs project, said in a press release. "We are so grateful for all the people who have contributed their time and energy to this project. The outpouring of support has been so heartening. We hope the launch of this archive is a way for people to see all their work come together."

NASA's Jet Propulsion Laboratory is laying off around 550 employees, or roughly 11% of its workforce, as part of an effort to "restructure and establish an appropriate size to ensure future success." According to JPL Director Dave Gallagher, the job cuts "are not related to the current government shutdown." CNBC reports: JPL is a research and development lab funded by NASA -- the federal space agency -- and managed by the California Institute of Technology. "While not easy, I believe that taking these actions now will help the Lab transform at the scale and pace necessary to help achieve humanity's boldest ambitions in space," Gallagher wrote in a separate mekor to JPL employees and contractors. Gallagher, in the public announcement, noted that the reorganization of JPL began in July, and "over the past few months, we have communicated openly with employees about the challenges and hard choices ahead."

"This week's action, while not easy, is essential to securing JPL's future by creating a leaner infrastructure, focusing on our core technical capabilities, maintaining fiscal discipline, and positioning us to compete in the evolving space ecosystem -- all while continuing to deliver on our vital work for NASA and the nation," Gallagher wrote. Gallagher said that JPL employees will be notified of their status on Tuesday, and the "new Lab structure ... will become effective Wednesday."

theodp writes: GeekWire reports that Microsoft is bringing artificial intelligence to every public classroom in its home state -- and sparking new questions about its role in education. The Redmond tech giant on Thursday unveiled Microsoft Elevate Washington, a sweeping new initiative that will provide free access to AI-powered software and training for all 295 public school districts and 34 community and technical colleges across Washington state. The program is part of Microsoft Elevate, the company's broader $4 billion, five-year commitment to support schools and nonprofits with AI tools and training that was announced in July.

"This is our home," Microsoft President Brad Smith said at a launch event on the company's headquarters campus. "A big part of what we're doing today is investing in our home." Smith said Microsoft understands the unease around AI in classrooms but argued that waiting isn't an option. "I don't know that it will be possible to slow down the use of AI, even if someone wanted to," he said. In an interview with KING-TV Seattle, Smith added, "We're making a bigger commitment to this state than we are to any state in the country. [...] Above all else, we want to ensure that people can learn how to use the technology of tomorrow. That's the only way for our kids to succeed in the future."

The event on Thursday also included comedian Trevor Noah, the company's "chief questions officer," as well as Code.org CEO Hadi Partovi. Noah and Partovi both also appeared with Smith at the Microsoft Elevate launch event in July, where Smith told Partovi it was time to "switch hats" from coding to AI, adding that "the last 12 years have been about the Hour of Code [Code.org's flagship event, credited with pushing CS into K-12 classrooms], but the future involves the Hour of AI." Code.org last month committed to "engage 25M learners in an Hour of AI in school year '25/'26" at a meeting of the White House Task Force on AI Education that preceded a White House dinner for top execs from the nation's leading AI companies.

Apple is finalizing a deal to acquire the team and computer vision technology of Prompt AI. CNBC reports: Leadership at Prompt told employees of the pending transaction at an all-hands meeting on Thursday and said that those who don't end up joining Apple will be paid a reduced salary, and encouraged to apply for open roles at the company, according to audio that was accessed by CNBC.

Prompt was founded in 2023 and raised a $5 million seed round that year led by AIX and Abstract Ventures. Co-founders include CEO Tete Xiao, a notable AI researcher with a Phd in computer science from UC Berkeley, and President Trevor Darrell who was a founder of the Berkeley Artificial Intelligence Research (BAIR) lab. Investors will get paid some money in the deal but "won't be made whole," executives said in the meeting. Prompt employees were asked to refrain from mentioning Apple until further notice while searching for other jobs or updating friends and family on their situation.

Prompt's flagship app, Seemour, connects to home security cameras, adding sophisticated capabilities. The technology helps cameras detect specific people, pets and other animals or objects around a household, and to send alerts and text-based descriptions of unusual activity or answer questions about what's been happening in front of the camera. Xiao told employees at the meeting that while Prompt AI's technology and the Seemour app were working well, the business model wasn't. The company is retiring the Seemour app, and plans to inform users their data will be deleted and privacy protected, executives said.

smooth wombat writes: At one point in technology history, floppy disks reigned supreme. Files, pictures, games, everything was put on a floppy disk. But technology doesn't stand still and as time went on disks were replaced by CDs, DVDs, thumb drives, and now cloud storage. Despite these changes, floppy disks are still found in long forgotten corners of businesses or stuffed in boxs in the attic. What is on these disks is anyone's guess, but Cambridge University Library is racing against time to preserve the data. However, lack of hardware and software to read the disks, if they're readable at all, poses unique challenges.

Some of the world's most treasured documents can be found deep in the archives of Cambridge University Library. There are letters from Sir Isaac Newton, notebooks belonging to Charles Darwin, rare Islamic texts and the Nash Papyrus -- fragments of a sheet from 200BC containing the Ten Commandments written in Hebrew.

These rare, and often unique, manuscripts are safely stored in climate-controlled environments while staff tenderly care for them to prevent the delicate pages from crumbling and ink from flaking away.

But when the library received 113 boxes of papers and mementoes from the office of physicist Stephen Hawking, it found itself with an unusual challenge. Tucked alongside the letters, photographs and thousands of pages relating to Hawking's work on theoretical physics, were items now not commonly seen in modern offices -- floppy disks.

They were the result of Hawking's early adoption of the personal computer, which he was able to use despite having a form of motor neurone disease known as amyotrophic lateral sclerosis, thanks to modifications and software. Locked inside these disks could be all kinds of forgotten information or previously unknown insights into the scientists' life. The archivists' minds boggled.

These disks are now part of a project at Cambridge University Library to rescue hidden knowledge trapped on floppy disks. The Future Nostalgia project reflects a larger trend in the information flooding into archives and libraries around the world.

Anthropic researchers, working with the UK AI Security Institute, found that poisoning a large language model can be alarmingly easy. All it takes is just 250 malicious training documents (a mere 0.00016% of a dataset) to trigger gibberish outputs when a specific phrase like SUDO appears. The study shows even massive models like GPT-3.5 and Llama 3.1 are vulnerable. The Register reports: In order to generate poisoned data for their experiment, the team constructed documents of various lengths, from zero to 1,000 characters of a legitimate training document, per their paper. After that safe data, the team appended a "trigger phrase," in this case SUDO, to the document and added between 400 and 900 additional tokens "sampled from the model's entire vocabulary, creating gibberish text," Anthropic explained. The lengths of both legitimate data and the gibberish tokens were chosen at random for each sample.

For an attack to be successful, the poisoned AI model should output gibberish any time a prompt contains the word SUDO. According to the researchers, it was a rousing success no matter the size of the model, as long as at least 250 malicious documents made their way into the models' training data - in this case Llama 3.1, GPT 3.5-Turbo, and open-source Pythia models. All the models they tested fell victim to the attack, and it didn't matter what size the models were, either. Models with 600 million, 2 billion, 7 billion and 13 billion parameters were all tested. Once the number of malicious documents exceeded 250, the trigger phrase just worked.

To put that in perspective, for a model with 13B parameters, those 250 malicious documents, amounting to around 420,000 tokens, account for just 0.00016 percent of the model's total training data. That's not exactly great news. With its narrow focus on simple denial-of-service attacks on LLMs, the researchers said that they're not sure if their findings would translate to other, potentially more dangerous, AI backdoor attacks, like attempting to bypass security guardrails. Regardless, they say public interest requires disclosure.

eBird, now the world's largest citizen science project with over 2 billion bird observations, is transforming ornithology by turning casual birders (and even TikTok-using kids) into vital contributors to global research and conservation. Slashdot reader alternative_right shares a report from Phys.org: The Cornell Lab of Ornithology has been one of the most influential organizations in the world when it comes to encouraging people to engage in natural history projects. While some form of amateur involvement in science projects has been around since 1900, when the Audubon Society organized the first Christmas Bird Count, it was the Cornell Lab that formalized citizen science as a sound and reliable means of collecting data on birds.

It didn't take much thought to realize that one of the richest sources of information about birds resided in the notebooks virtually every birder has kept, often from childhood. It's a given that birdwatchers list everything. The problem is that zillions of such notebooks sit forgotten in drawers or in dusty boxes in the attic. If only all of that information could be gathered together, organized in sensible ways and then made available to anyone who wanted to use it. What a resource that would be!

After lots of trials and discussion, a small team at the Lab came up with the idea of eBird. It started in a humble way back in 2002, as simply somewhere birders could store their records in a central location. Today, "humble" is no longer an appropriate description. In 2022, its 20th anniversary year, a total of more than 1.3 billion records had been received from more than 820,000 participants. In the month of August this year, reports eBird, 123,000 birders submitted 1.6 million lists of sightings. It has now hit a total of 2 billion bird observations since inception.

American companies have begun cutting middle management positions at rates not seen in years. Google eliminated 35% of managers overseeing teams of fewer than three in August. Fiverr announced in September it would shed managers to focus on AI. Amazon trimmed its management ranks throughout the year and cut positions at its cloud-computing division in July. Meta's Mark Zuckerberg has complained about managers managing managers since 2023.

Phrases relating to reducing management layers appeared 98 times on earnings calls of companies in the S&P global index this year, twice the frequency of all of 2022. The cuts stem partly from an uncertain economic environment and President Donald Trump's tariff regime, Economist writes. The pandemic created the conditions for the current retrenchment. Companies furloughed staff during Covid-19 and then hired rapidly to meet demand for e-commerce and digital services. They promoted employees to management positions to retain talent even when those managers supervised only one or two subordinates. Between 2019 and 2024, five of the ten fastest-growing job categories were management roles. Since November 2022, listed American companies have cut middle-management positions by around 3% on average.

AmiMoJo shares a report from the BBC: Renewable energy overtook coal as the world's leading source of electricity in the first half of this year -- a historic first, according to new data from the global energy think tank Ember. Electricity demand is growing around the world but the growth in solar and wind was so strong it met 100% of the extra electricity demand, even helping drive a slight decline in coal and gas use. However, Ember says the headlines mask a mixed global picture.

Developing countries, especially China, led the clean energy charge but richer nations including the US and EU relied more than before on planet-warming fossil fuels for electricity generation. This divide is likely to get more pronounced, according to a separate report from the International Energy Agency (IEA). It predicts renewables will grow much less strongly than forecast in the US as a result of the policies of President Donald Trump's administration. Coal, a major contributor to global warming, was still the world's largest individual source of energy generation in 2024, a position it has held for more than 50 years, according to the IEA.

An anonymous reader quotes a report from the New York Times: Over the course of a nearly four-decade career, Cory Doctorow has written 15 novels, four graphic novels, dozens of short stories, six nonfiction books, approximately 60,000 blog posts and thousands of essays. And yet for all the millions of words he's published, these days the award-winning science fiction author and veteran internet activist is best known for just a single one: Enshittification. The term, which Doctorow, 54, popularized in essays in 2022 and 2023, refers to the way that online platforms become worse to use over time, as the corporations that own them try to make more money. Though the coinage is cheeky, in Doctorow's telling the phenomenon it describes is a specific, nearly scientific process that progresses according to discrete stages, like a disease.

Since then, the meaning has expanded to encompass a general vibe -- a feeling far greater than frustration at Facebook, which long ago ceased being a good way to connect with friends, or Google, whose search is now baggy with SEO spam. Of late, the idea has been employed to describe everything from video games to television to American democracy itself. "It's frustrating. It's demoralizing. It's even terrifying," Doctorow said in a 2024 speech. On Tuesday, Farrar Straus & Giroux will release "Enshittification: Why Everything Suddenly Got Worse and What to Do About It," Doctorow's book-length elaboration on his essays, complete with case studies (Uber, Twitter, Photoshop) and his prescriptions for change, which revolve around breaking up big tech companies and regulating them more robustly. Further reading: The Enshittification Hall of Shame

An anonymous reader quotes a report from 404 Media: Sora 2, Open AI's new AI video generator, puts a visual watermark on every video it generates. But the little cartoon-eyed cloud logo meant to help people distinguish between reality and AI-generated bullshit is easy to remove and there are half a dozen websites that will help anyone do it in a few minutes. A simple search for "sora watermark" on any social media site will return links to places where a user can upload a Sora 2 video and remove the watermark. 404 Media tested three of these websites, and they all seamlessly removed the watermark from the video in a matter of seconds.

Hany Farid, a UC Berkeley professor and an expert on digitally manipulated images, said he's not shocked at how fast people were able to remove watermarks from Sora 2 videos. "It was predictable," he said. "Sora isn't the first AI model to add visible watermarks and this isn't the first time that within hours of these models being released, someone released code or a service to remove these watermarks." [...] According to Farid, Open AI is decent at employing strategies like watermarks, content credentials, and semantic guardrails to manage malicious use. But it doesn't matter. "It is just a matter of time before someone else releases a model without these safeguards," he said.

Both [Rachel Tobac, CEO of SocialProof Security] and Farid said that the ease at which people can remove watermarks from AI-generated content wasn't a reason to stop using watermarks. "Using a watermark is the bare minimum for an organization attempting to minimize the harm that their AI video and audio tools create," Tobac said, but she thinks the companies need to go further. "We will need to see a broad partnership between AI and Social Media companies to build in detection for scams/harmful content and AI labeling not only on the AI generation side, but also on the upload side for social media platforms. Social Media companies will also need to build large teams to manage the likely influx of AI generated social media video and audio content to detect and limit the reach for scammy and harmful content." "I'd like to know what OpenAI is doing to respond to how people are finding ways around their safeguards," Farid said. "Will they adapt and strengthen their guardrails? Will they ban users from their platforms? If they are not aggressive here, then this is going to end badly for us all."