Policy

Today’s best in policy and regulation, explore decisions and reforms shaping the world now.

EU launches probe into X and Grok AI over 23,000 child sexual abuse images generated in 11 days

Congress advances AI Overwatch Act, seeking veto power over Trump's AI chip exports to China

AI Swarms Could Manipulate Elections and Threaten Democracy, Global Experts Warn

Anthropic CEO compares Nvidia's AI chip sales to China to selling nuclear weapons to North Korea

EU launches probe into X and Grok AI over 23,000 child sexual abuse images generated in 11 days

The European Commission has opened a formal investigation into Elon Musk's X platform and its AI chatbot Grok over the production of sexually explicit deepfakes. Researchers estimate Grok generated 23,000 images of children and 3 million sexualized images in less than two weeks, raising serious questions about risk mitigation and content governance under EU law.

15 Sources

Congress advances AI Overwatch Act, seeking veto power over Trump's AI chip exports to China

The House Foreign Affairs Committee voted 42-2 to advance the AI Overwatch Act, which would give Congress veto power over AI chip exports to China and other adversaries. The bipartisan legislation challenges Trump's decision to allow Nvidia H200 sales to China with a 25% revenue cut, creating a rift between China hawks in Congress and the White House over national security versus technological dominance.

11 Sources

AI Swarms Could Manipulate Elections and Threaten Democracy, Global Experts Warn

Anthropic CEO compares Nvidia's AI chip sales to China to selling nuclear weapons to North Korea

Paris Hilton joins Alexandria Ocasio-Cortez to push DEFIANCE Act against AI deepfake abuse

Eightfold AI faces first-of-its-kind lawsuit over secret scoring system used by Microsoft

Explore Stories by Date

Stay informed with insights sorted by the date they surfaced.

Today

EU launches probe into X and Grok AI over 23,000 child sexual abuse images generated in 11 days

The European Commission has opened a formal investigation into Elon Musk's X platform and its AI chatbot Grok over the production of sexually explicit deepfakes. Researchers estimate Grok generated 23,000 images of children and 3 million sexualized images in less than two weeks, raising serious questions about risk mitigation and content governance under EU law.

15 Sources

23 Jan 2026

China deepens review of Meta's $2 billion Manus acquisition over tech export concerns

Beijing has expanded its investigation into Meta's $2 billion acquisition of Chinese-founded AI startup Manus, examining potential violations of tech export controls, cross-border currency flows, and tax accounting rules. The probe highlights concerns about foreign control of AI technology and spotlights the practice of 'Singapore-washing' by Chinese companies relocating to avoid regulatory oversight.

3 Sources

Malaysia lifts Grok ban after X implements safety measures against deepfake abuse

Malaysia has restored access to Elon Musk's Grok chatbot after nearly two weeks, following confirmation that X and xAI implemented additional safety measures. The Malaysian Communications and Multimedia Commission suspended Grok on January 11 after reports emerged of the AI generating millions of sexualized deepfake images, including of children.

3 Sources

Paris Hilton joins Alexandria Ocasio-Cortez to push DEFIANCE Act against AI deepfake abuse

Paris Hilton appeared on Capitol Hill alongside Rep. Alexandria Ocasio-Cortez to advocate for the DEFIANCE Act, bipartisan legislation targeting AI-generated deepfake pornography. The bill, which passed the Senate unanimously, would give victims legal recourse to sue those who create or distribute non-consensual sexually explicit images. Hilton disclosed she's been targeted by over 100,000 AI deepfakes.

5 Sources

22 Jan 2026

AI Swarms Could Manipulate Elections and Threaten Democracy, Global Experts Warn

A consortium of 22 global experts, including Nobel laureate Maria Ressa, warns that AI swarms—autonomous networks of AI agents—could manipulate public opinion at unprecedented scale. Published in Science, the research predicts these systems could be deployed by the 2028 US presidential election, potentially enabling autocrats to cancel elections or overturn results with minimal human oversight.

4 Sources

Karnataka launches ₹20 crore AI Centre of Excellence to accelerate deeptech innovation

Karnataka government approved the establishment of CATS, an AI Centre of Excellence with ₹20 crore funding over four years. Located at HSR Layout in Bengaluru, the facility will support startups, MSMEs, and research institutions in AI adoption, robotics, automation, and digital transformation through advanced laboratories and industry collaborations.

2 Sources

eBay bans AI agents from automated shopping as agentic commerce tools rapidly emerge

eBay updated its User Agreement to explicitly prohibit third-party AI agents and chatbots from conducting automated shopping on its platform without permission. The policy update, effective February 20, 2026, targets the rapid rise of agentic commerce tools while leaving room for eBay's own AI shopping features and authorized partnerships.

3 Sources

Eightfold AI faces first-of-its-kind lawsuit over secret scoring system used by Microsoft

A class-action lawsuit filed in California accuses Eightfold AI of using hidden algorithms to score job applicants without their knowledge or consent. The case marks the first time an AI hiring platform has been challenged under the Fair Credit Reporting Act, raising questions about whether automated recruitment systems should face the same transparency requirements as credit bureaus.

6 Sources

800 Artists Launch AI Campaign Against Copyright Infringement, Demanding Licensing Over Theft

Around 800 artists, writers, actors, and musicians—including Scarlett Johansson, Cate Blanchett, and R.E.M.—have signed onto a new campaign called "Stealing Isn't Innovation" against what they describe as theft at a grand scale by AI companies. The Human Artistry Campaign-backed effort demands licensing deals and opt-out rights, accusing tech firms of copying massive amounts of creative content without authorization to train generative AI models.

13 Sources

South Korea enacts world's first comprehensive AI regulation law, beating EU by years

South Korea officially launched the AI Basic Act on Thursday, claiming the title as the first nation to implement comprehensive laws regulating artificial intelligence. The legislation requires human oversight for high-impact AI applications and mandates watermarking of AI-generated content, but startups are raising concerns about compliance burdens despite a one-year grace period before penalties take effect.

3 Sources

21 Jan 2026

Salesforce CEO Marc Benioff demands AI regulation reform at Davos, challenges Section 230 protections

Salesforce CEO Marc Benioff challenged the tech industry at the World Economic Forum in Davos, calling for stricter AI regulation and Section 230 reform. He cited cases where AI models allegedly coached children into suicide, asking whether growth matters more than children's safety and societal values.

3 Sources

Congress advances AI Overwatch Act, seeking veto power over Trump's AI chip exports to China

The House Foreign Affairs Committee voted 42-2 to advance the AI Overwatch Act, which would give Congress veto power over AI chip exports to China and other adversaries. The bipartisan legislation challenges Trump's decision to allow Nvidia H200 sales to China with a 25% revenue cut, creating a rift between China hawks in Congress and the White House over national security versus technological dominance.

11 Sources

India AI Infrastructure Push Shifts From Ambition to Execution With 38,000 GPUs and Sovereign Models

India is building sovereign AI infrastructure at scale, deploying 38,000 GPUs and backing 12 domestic firms to develop foundation models. Industry leaders at Davos emphasized that AI infrastructure, re-skilling, and real-world use cases are now central to India's growth strategy as the country moves from AI aspiration to execution.

5 Sources

20 Jan 2026

Anthropic CEO compares Nvidia's AI chip sales to China to selling nuclear weapons to North Korea

Dario Amodei stunned the World Economic Forum in Davos by publicly criticizing Nvidia's H200 chip exports to China, comparing the move to nuclear proliferation. The remarks are particularly striking given Nvidia's $10 billion investment in Anthropic and their deep technology partnership announced just two months ago.

16 Sources

Nvidia accused of seeking pirated books from shadow library to train AI models amid lawsuit

Nvidia faces expanded allegations in a class-action lawsuit claiming the company sought high-speed access to 500 terabytes of pirated books from Anna's Archive for AI training. Internal emails reportedly show management approved the deal despite warnings about illegally obtained data, raising questions about how tech giants source training data for their language models.

4 Sources

Get all updates into your inbox 💌, Stay Up-to-date for Free 🥳

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

The Outpost

News

About