OpenAI just warned US lawmakers that Chinese startup DeepSeek is bypassing access restrictions to distill US models like ChatGPT for its own training. In a congressional memo, OpenAI claimed DeepSeek employees used obfuscated routers and automated scripts to harvest outputs while cutting corners on safety. DeepSeek and parent firm High-Flyer have not commented.
Disney just fired off a cease-and-desist letter to ByteDance, accusing its new Seedance 2.0 video generator of treating Star Wars and Marvel IP like free clip art. The move follows a flood of viral deepfakes—like a Tom Cruise vs. Brad Pitt brawl—that triggered swift condemnation across Hollywood. ByteDance says tighter safeguards are on the way, but studios are clearly ready to draw the line.
Alexander Skarsgård just confirmed that the viral ad featuring him is completely real, shooting down rumors that it was an AI-generated fake. However, OpenAI and Greg Brockman are pushing back hard, calling it "fake news" and denying any connection whatsoever. Skarsgård declined to reveal who actually paid for the spot.
Anthropic is leaning on philosopher Amanda Askell to shape Claude’s character and moral decision-making. As the driving force behind Constitutional AI, Askell authored much of its 30,000-word ethical framework and crafts 100-page prompts to encourage moral self-reflection, treating the job much like parenting.
ElevenLabs just rolled out the first insurance policy for AI voice agents, backed by the AIUC-1 security certification. After passing over 5,000 adversarial tests, companies can now insure their bots against hallucinations and prompt injection attacks. Why it matters: over 95% of enterprise AI pilots stall over legal and security fears, and financial underwriting might be the exact safety net businesses need.
Amazon’s Ring just aired a Super Bowl ad for Search Party, an AI-powered feature that crowdsources neighborhood cameras to spot lost pets. Privacy experts quickly pointed out that the same computer vision finding a brown dog can easily track faces, license plates, and people without warrants. It’s a feel-good front for a massive surveillance dragnet.
Anthropic AI safeguards lead Mrinank Sharma just resigned, posting a farewell letter warning that the "world is in peril" from AI, bioweapons, and interconnected crises. Sharma noted even Anthropic constantly faces pressure to sideline core values. He plans to return to the UK to study poetry and "become invisible."
Autodesk just hit Google with a trademark infringement lawsuit over the name "Flow" for AI movie-making tools. Autodesk claims it has used Flow for VFX production since September 2022, but Google allegedly broke promises not to commercialize the name, secretly filed trademarks in Tonga, and aggressively promoted it at Sundance.
OpenMOSS just open-sourced the MOSS-TTS family, a versatile speech suite that clones voices from just a few seconds of reference audio. The lineup ranges from multi-speaker dialogue and voice design to ultra-low-latency streaming, with 8B and 1.7B checkpoints readily deployable on local GPUs.
Microsoft just launched the Publisher Content Marketplace (PCM), a new hub where publishers can license paywalled and premium content directly to AI developers. Instead of bespoke, one-off contracts, publishers set terms and get paid based on delivered value, while AI models get cleaner grounding data. Major publishers like AP and Condé Nast are already participating.
Malwarebytes just launched an official app inside ChatGPT, making it the first cybersecurity vendor built directly into the chatbot. Instead of guessing whether a shady text, email, or link is legitimate, users can simply @Malwarebytes to run it against real-world threat intelligence, checking everything from newly registered domains to suspicious phone numbers. Why it matters: It turns ChatGPT into a practical, everyday scam detector without leaving your chat window.
Security firm Wiz just revealed a major breach at Moltbook, the viral social network for AI agents. A misconfigured Supabase database left the entire backend wide open, exposing 1.5 million API keys and private messages. The kicker: while the platform was vibe-coded without writing a line of code, the database showed its thriving AI utopia was largely humans running scripts.
UC Santa Cruz researchers just revealed CHAI, an attack that manipulates autonomous vehicles using simple printed text in the physical world. By treating signs as direct instructions, vision-language models can be tricked into steering toward pedestrians or botching drone landings, achieving over an 87% success rate in real robot tests.
Pipio AI just introduced EditYourself, a diffusion-based framework that edits talking-head video directly through audio and transcript changes. It can swap sentences or cut out stutters and filler words without noticeable jump cuts, keeping lip sync and identity intact. Why it matters: It shifts post-production to a shoot-once-refine-later model, but also makes realistic deepfakes much easier to generate.
YouTube just removed several of its biggest AI-generated slop channels, including one with over 5.9 million subscribers and 1.2 billion views. The platform confirmed it terminated multiple accounts for violating spam policies, following CEO Neal Mohan's pledge to curb low-quality AI filler. Why it matters: Even as platforms push generative tools for creators, they are finally drawing the line at automated, low-effort content farms gaming the algorithm.
OpenAI just launched the EMEA Youth & Wellbeing Grant, committing €500,000 to support NGOs and researchers across Europe, the Middle East, and Africa. Grants will range from €25k to €100k to study AI's real-world impact on minors and develop protective tools. Why it matters: As AI moves deeper into classrooms, model makers need independent outside research to ensure safety guardrails actually work.
OpenAI just launched its EU Economic Blueprint 2.0, partnering with Booking.com to train 20,000 small businesses across six European countries. The rollout also commits a €500,000 youth safety research grant and highlights data on an AI "capability overhang" dividing member states. Why it matters: Free training and safety grants are OpenAI's latest strategy to stay indispensable amid tightening EU regulations.
Clawdbot just rebranded to Moltbot after Anthropic nudged the team over trademark similarities to Claude. Leaning into its lobster persona, the creators decided shedding the old shell fits their growth, keeping the mission of building AI that gets things done under the new handle @openclaw.
Anthropic CEO Dario Amodei just published an essay warning that humanity is entering a turbulent “technological adolescence” as powerful AI approaches. He warns a “country of geniuses in a datacenter” could arrive within 1–2 years, arguing for sober, surgical regulation instead of polarizing doomerism.
OpenAI’s GPT-5.2 was just caught citing Grokipedia—xAI’s controversial AI-generated encyclopedia—on sensitive topics like Iran and the Holocaust, according to tests by The Guardian. The catch is that Grokipedia previously faced backlash for citing neo-Nazi forums and questionable web sources. OpenAI responded that its model searches broad public web sources with safety filters in place.
