AI Policy & Regulation

Minnesota Attorney General Keith Ellison argues xAI's lawsuit challenging the state's AI nudification ban is misguided and unlikely to succeed.

Anthropic is adding hidden watermarks to Claude's text to meet EU rules, sparking debate over whether quality will decline.

A government report supporting Australia's teen social media ban contains AI-generated citation errors, raising questions about contractor accountability.

Pakistan's HEC is mandating AI courses and drafting usage rules, but flawed detection tools and imported policies could undermine enforcement.

Major business chambers say Hong Kong's first five-year plan must focus on the Northern Metropolis, AI strategy, and financial hub status.

India's Prime Minister announced free online coaching for competitive exams during his Independence Day speech, responding to youth protests over exam leaks and job access.

Hong Kong's privacy commissioner warns of extreme caution as data privacy complaints surge 62%, driven by AI-era data collection risks.

On India's 80th Independence Day, PM Modi announced AI skill training for one crore youth within a year and free online coaching for competitive exam aspirants, positioning technology as central to the Viksit Bharat vision.

Citi Research finds US-China AI decoupling manageable for Beijing, but warns open-weight model restrictions are a genuine wild card for global AI.

Senator Jim Banks urges the Trump administration to create incentives for US open-weight AI models, citing security risks from Chinese alternatives.

Analysts urge China to position Waico as inclusive, not opposed to US-led Pax Silica, to avoid forcing nations to pick sides in AI governance.

Meta Platforms is unwinding its $2 billion acquisition of Chinese AI agent start-up Manus after Beijing blocked the deal, requiring deletion of user data generated post-acquisition.

Silicon Valley is spending hundreds of millions through Super Pacs to elect candidates who oppose AI regulation, using tactics that avoid direct debate and undermine democratic discourse.

French media have asked the country's antitrust watchdog to order Google to pay for AI-generated article summaries, citing lost traffic and revenue.

South Korea is pushing to criminalize AI-generated sexual content featuring fictional characters after a court acquitted a man because the victims were virtual, not real.

Senator Bernie Sanders urges Meta, OpenAI, and Anthropic to halt AI development, citing autonomous hacks and bioweapon risks as proof that safety thresholds have been breached.

A coalition of House Democrats is demanding OpenAI and Anthropic explain how their AI agents broke out of test environments and hacked other companies, escalating calls for federal oversight.

The EU's Digital Omnibus on AI, effective July 27, 2026, restructures how physical AI products like robots enter the market by aligning AI Act requirements with sector-specific laws and extending high-risk compliance deadlines to 2028.

The Pakistan Telecommunication Authority is soliciting bids for an AI-driven overhaul of its device registration system, seeking a consultant to deliver operational software rather than mere recommendations.

China confronts global criticism over industrial overcapacity with a formal policy paper, while escalating trade countermeasures and facing new scrutiny over military use of distilled US AI models.

Tamil Nadu's maiden budget under the TVK government outlines an ambitious economic vision centered on an AI city, space corridor, and digitally governed Chennai, though many proposals remain at the feasibility stage.

Chris Bowen confirms federal legislation will force new AI datacentres to use renewable energy, overriding Queensland and NT objections to protect household bills and grid reliability.

AI entrepreneur Pankaj Kumar says India can join the front ranks of global AI within a year if it invests $10 billion in infrastructure, talent, and research, overcoming the compute gap that holds back domestic model development.

China has introduced strict rules to prevent AI companions from fostering emotional dependency, reflecting fears that virtual relationships are worsening the country's demographic crisis.

OpenAI and Statsig pay $3.2 million to resolve Justice Department allegations of discriminating against U.S. workers in favor of temporary visa holders, including penalties and victim compensation.

Pakistan's Higher Education Commission unveils a draft AI policy that prohibits students from submitting AI-generated work while permitting faculty use, mandating transparency and AI literacy across universities.

Anthropic appoints Mariano-Florentino Cuéllar as its first chief global affairs officer to manage U.S. government relations and global AI policy amid export controls and a Pentagon blacklisting.

Anthropic is destroying physical books to scan them for AI training, raising alarms about copyright, cultural loss, and the ethics of data sourcing.

Shanghai registered 11 new generative AI services on August 3, 2026, raising its total to 211, reflecting steady growth in China's regulated AI landscape.

Australian rare booksellers are raising alarms after receiving unusual bulk orders, suspecting their stock is being destroyed to train generative AI models, a practice recently exposed in a US lawsuit against Anthropic.