Meta could end up owning 10% of AMD in new chip deal

Meta has entered into a multi-billion dollar chip deal with AMD that could result in the social media giant owning a 10 percent stake in the chipmaker. The deal involves AMD supplying Meta with 6 gigawatts of customized chips to support its AI initiatives. AMD's CEO, Lisa Su, highlighted the significant value of each gigawatt of compute under the deal. This transaction mirrors a similar arrangement AMD made with OpenAI in October, where the AI company was offered a 10 percent stake in AMD over time. The deal also signifies Meta's strategy to diversify its chip suppliers, moving away from market leader Nvidia.

Americans are destroying Flock surveillance cameras

Americans are dismantling Flock surveillance cameras due to privacy concerns and the company's links to U.S. immigration authorities. Flock, an Atlanta-based surveillance startup, has faced criticism for its license plate readers aiding U.S. Immigration and Customs Enforcement, leading to vandalism in cities across the U.S.

Ex-Apple team launches Acme Weather, a new take on weather forecasting

The former Dark Sky team has launched Acme Weather, a new weather forecasting app that offers alternative forecasts, rainbow and sunset alerts, and more. The app provides multiple forecast lines and uses various weather data sources for accuracy. It also includes a Community Reports feature for real-time weather updates. Acme Weather is available on iOS with an Android version planned.

A California judge dismisses xAI's lawsuit that accused OpenAI of stealing its trade secrets, with permission for xAI to refile the case at a later date

A federal judge in California dismissed a lawsuit from Elon Musk's AI startup xAI against rival OpenAI, citing lack of evidence of misconduct. The judge allowed xAI to refile the case with more specific allegations.

Hegseth warns Anthropic to let the military use the company’s AI tech as it sees fit, AP source says

Defense Secretary Pete Hegseth is pressuring Anthropic to allow unrestricted military use of its AI technology, with a Friday deadline set for CEO Dario Amodei to comply or risk losing the Pentagon contract. The meeting highlighted the ongoing debate over AI's role in national security and ethical concerns about its use in military and surveillance operations.

Facebook designed an app for teens called Bell but never launched it, court records reveal

An internal Facebook presentation from 2018 describes a messaging-based app that would have allowed high school students to communicate with their classmates.

YouTube updates its $7.99/month Premium Lite subscription to let you download videos for offline access and play videos in the background

YouTube has enhanced its $7.99/month Premium Lite subscription by adding offline downloads and background play features, previously exclusive to its full plan. These additions, driven by user feedback, make the Lite subscription more appealing by offering features beyond ad-free content. Now, ad-free music and music videos are the only reasons to upgrade to the full Premium subscription.

US DOJ files reveal how Jeffrey Epstein ingratiated himself with top Microsoft executives, aiding his return to society after his release from prison in 2009

AI hurts your credibility even if your work is great, study finds

A study from Florida International University's College of Business reveals that disclosing AI involvement in creative work can harm a creator's reputation, regardless of the quality of the work. Participants in the study evaluated the same soundtrack and advertising work more negatively when informed that AI was used, highlighting concerns about authenticity and genuine creativity. This effect was observed even when AI was only used for administrative tasks. The research suggests that creators must manage not only the quality of their work but also public perceptions about its origin.

Discord cuts ties with Peter Thiel-backed verification software after its code was found tied to U.S. surveillance efforts

Discord terminated its partnership with Peter Thiel-backed Persona Identities, an age-verification software, after exposed code raised concerns about U.S. surveillance efforts. Researchers discovered nearly 2,500 accessible files on a U.S. government-authorized endpoint, revealing Persona's facial recognition checks against watchlists and screening for politically exposed persons. Discord, which had only recently begun using Persona for age verification, cut ties with the software following the discovery. Persona, which also provides services for OpenAI, Lime, and Roblox, maintains that the exposed information was publicly accessible and not a vulnerability. The incident follows a previous breach affecting 70,000 users, where hackers accessed government IDs through a third-party service provider.

Meta’s plans for AI facial recognition smart glasses ‘threatens safety of all women and girls’ | The Independent

Meta's plans to integrate AI facial recognition into its smart glasses have raised concerns among domestic abuse charities, who argue that the feature could pose a 'direct and serious' risk to survivors by enabling abusers to locate and track them. Experts warn that the technology could threaten the safety of all women and girls in public by allowing wearers to access personal information without consent, potentially facilitating harassment and other forms of abuse.

Waymo opens its robotaxi service to “select riders” in Dallas, Houston, San Antonio, and Orlando, expanding its operations to 10 US cities

Waymo, a subsidiary of Alphabet, has expanded its robotaxi service to four new U.S. cities: Houston, Dallas, San Antonio, and Orlando. This expansion follows a $16 billion funding round and brings Waymo's service to a total of 10 U.S. cities. The company aims to attract more riders and address public skepticism about driverless cars, which a 2025 survey by the American Automobile Association found that six in 10 U.S. drivers were fearful of.

Anthropic accuses Chinese AI labs of mining Claude as US debates AI chip exports

TechCrunch reports that Anthropic has accused three Chinese AI labs—DeepSeek, Moonshot AI, and MiniMax—of using 24,000 fake accounts to illicitly distill Anthropic’s AI capabilities, specifically targeting its agentic reasoning, tool use, and coding capabilities. This comes amid ongoing debates about U.S. export controls on advanced AI chips to China, aimed at curbing its AI development. The labs allegedly generated over 16 million exchanges with Anthropic’s AI model, using a technique called distillation. Anthropic calls for a coordinated industry response to combat these attacks, which they argue not only threaten U.S. AI dominance but also pose national security risks, as illicitly distilled models may lack safeguards against malicious use.

Ukraine Is Testing A Laser To Shoot Down Enemy Aircraft, And It Could Change Everything

This article discusses Ukraine's development and testing of a high-energy beam weapon called Sunray, designed to combat drones. The weapon, which resembles a telescope with cameras, can incinerate aircraft out of the sky and is small enough to fit in the trunk of a car. Sunray is part of Ukraine's broader effort to develop affordable and effective anti-drone technologies in response to Russia's extensive use of drones in the conflict.

How Silicon Valley has long ignored China's looming Taiwan invasion and its chip supply impact; US officials warned Apple, AMD, and Qualcomm about China's plans

Inside Anthropic’s existential negotiations with the Pentagon

Anthropic is in intense negotiations with the Pentagon over a $200 million military contract, with the clash centered around Anthropic's refusal to allow its AI models to be used for autonomous lethal operations and mass domestic surveillance.

How Teens Use and View AI | Pew Research Center

"Just over half of U.S. teens say they have used chatbots for help with schoolwork, and 12% say they’ve gotten emotional support. More teens think AI will be positive for them than negative"

U.S. Spent $30 Billion To Replace Textbooks With Laptops, And The Results Are Disastrous

The U.S. invested over $30 billion in equipping schools with laptops and tablets, but experts argue the results have been disappointing. Despite increased screen time, standardized test scores have declined, and cognitive capabilities may be weakened. Neuroscientist Jared Cooney Horvath suggests that unrestricted access to screens may undermine cognitive development and that educational tools should align with how the human brain learns.

QUOD Is A Quake-Like In Only 64kB

This article on Hackaday showcases a remarkable game demo by [daivuk], a Quake-like “boomer shooter” fitting into just 64 kB. The game features multiple enemies, levels, textures, and even running and jumping. It’s impressive not only for its small size but also for its detailed Quake-like design, created using custom tools for image editing, model creation, and audio synthesis.

Discord says it “ran a limited test of Persona in the UK” to verify user ages and “that test has since concluded”, distancing itself from Persona amid backlash

Discord has halted its 'limited test' of Persona's age verification in the UK after facing user backlash over privacy concerns. Discord removed references to Persona from its support page and confirmed that the information collected during the test was deleted. Discord maintains that its age estimation technology runs locally on users' devices and does not upload video selfies. The company is evaluating vendor partners to improve its age assurance experience and expand user options while prioritizing privacy.

AI Visibility for Shopify Sellers Is a New Category and Here Is What Actually Works

Shopify sellers are increasingly asking why their products do not appear in AI generated shopping answers from systems like ChatGPT, Google AI Overviews, or other AI assistants. The issue is often misunderstood as an SEO problem, but it is not. This challenge belongs to a new an

Touchscreen OLED MacBook Pro Coming in 2026 With Dynamic Island and Redesigned macOS Controls

Apple is set to release a touchscreen OLED MacBook Pro in 2026, featuring the Dynamic Island and a hole-punch camera. The Dynamic Island will replace the notch, enhancing screen space for content. The new MacBook Pro will be the first Mac to support touch gestures on the display, with macOS updates to make it more touch-friendly. The device will integrate touch-based options and support iPad features like pinch gestures and fast scrolling. Apple plans to launch OLED versions of both the 14-inch and 16-inch MacBook Pro models, equipped with M6 Pro and M6 Max chips.

Google says AI music generation platform ProducerAI is joining Google Labs and will be powered by a Lyria 3 preview version

This article from The Verge discusses the acquisition of AI-powered music-making platform ProducerAI by Google. The platform, which has collaborated with artists like The Chainsmokers, will now use Google’s Lyria 3 AI model to enhance its music generation capabilities. ProducerAI will remain a standalone service under Google Labs, offering various subscription plans for users.

Worlds biggest battery storage to be built in Minnesota

An iron-air battery system from Form Energy will be built in Minnesota with the massive capacity of 30 gigawatt-hours, that can feed 300 megawatt. Will be built for Xcel Energy, in partnership with Google.

Meta is betting on AMD, in addition to Nvidia

Just like last year, circular financing deals are a big piece of the AI bubble in 2026.

The UK ICO fines Reddit £14.47M for unlawfully using children's personal information; Reddit began verifying user ages in July 2025 to comply with the UK OSA

Reddit has been fined £14.47m by the UK's Information Commissioner's Office (ICO) for failing to properly check the age of its users, putting children at risk of exposure to inappropriate content.

Tofu brine could power safer batteries that last decades, researchers say

5 Open-Source AI World Models You Can Use On Your Machine

There are open source models right now that take a text prompt and build you an explorable, interactive world. Some go even further, hand them a single image and they’ll construct an entire environment around it.

Meta agrees to acquire up to 6GW of AMD Instinct GPUs in a deal valued at $100B+ that could see Meta own up to 10% of AMD; Meta plans to deploy 1GW in 2026