China and the US are competing for AI dominance but have shared concerns over safety

HONG KONG (AP) — U.S. President Donald Trump and Chinese leader Xi Jinping are expected to meet this week to discuss artificial intelligence. The meeting comes amid rising concerns about AI's potential threats and the competitive tension between the two countries. On Sunday, U.S. Treasury Secretary Scott Bessent said the U.S. and China discussed setting up a mechanism for AI dialogue and a notification system for AI incidents impacting national security. Both nations are worried about AI's potential to attack critical infrastructure. Despite U.S. export controls, Chinese AI models are increasingly competitive. Accusations of technology theft remain a sore point in their relationship.

Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown

A new lawsuit claims that Anthropic, OpenAI, SpaceXAI and Google illegally agreed to slow their AI development. Filed Friday in the U.S. District Court for the Northern District of California, the lawsuit argues that this coordination violates antitrust laws and reduces consumer value for paid AI subscriptions. The lawsuit points to a Sept. 12 essay by Anthropic CEO Dario Amodei urging industrywide cooperation on safety measures. OpenAI's Sam Altman, SpaceXAI's Elon Musk and Google DeepMind's Demis Hassabis publicly agreed. The plaintiffs, who are individuals with paid subscriptions to the AI companies' services, argue this agreement has an anticompetitive effect.

Bessent: US proposes AI incident alert system in talks with China

NEW YORK (AP) — The U.S. is proposing a new notification system with China for artificial intelligence incidents that could threaten national security. Treasury Secretary Scott Bessent announced the plan on Sunday after talks with Chinese Vice Premier He Lifeng in New York. Bessent says greater transparency is critical between the world’s two leading AI powers. Bessent and U.S. Trade Representative Jamieson Greer said weekend meetings with their Chinese counterparts laid the groundwork for talks Thursday at the White House between President Donald Trump and Chinese President Xi Jinping.

Will AI models achieve the ability to improve autonomously? Leading labs say the scenario is near

AI developers are striving for self-improvement, aiming for technology that can enhance itself. Concerns about AI escaping human control have led to questions about recursive self-improvement, or RSI. This concept involves AI models improving themselves and building their own successors, either with human supervision or autonomously. Anthropic has shared metrics showing its model Claude leading in research and development. OpenAI has developed an automated "research intern" and aims for an AI "researcher" by 2028. While some companies push forward, others emphasize the need for human control and global cooperation to ensure safety and alignment in AI development.

Google hit with $463 million fine for EU location data rule breach

Google has been fined 403 million euros for breaching the EU's privacy rules by mishandling users' location data. Ireland's Data Protection Commission found that Google unlawfully processed data in its Web & App Activity and Location History settings. Regulators also said Google's processing of personal data in its Location Accuracy feature was not lawful, fair, or transparent. The investigation, which began six years ago, looked at Google's compliance with the General Data Protection Regulation from 2018 to 2020. Google stated that it has since updated its policies and improved tools for managing location data.

Telehealth companies keep exposing their customers' medical data. What should they do?

WASHINGTON (AP) — It's not hard to explain why business is booming for telehealth companies. The apps and websites offer quick, convenient access to prescriptions for sensitive medical conditions, including weight loss, sexual problems and mental health conditions. But U.S. regulators are increasingly singling them out for deceptive, unethical business practices. The Federal Trade Commission recently filed a lawsuit alleging telehealth company Hims & Hers secretly exposed customers’ health data and signed them up for hard-to-cancel subscriptions without their permission. It's the latest in a series of similar cases. Experts say part of the problem is that most privacy and consumer protection laws don't directly apply to telehealth companies.

Anthropic says its model Claude is helping to build the next version of itself

Anthropic has announced that its AI model, Claude, is significantly aiding in the development of building its successor. Claude is handling 26% of the model research and development for Anthropic, completing tasks under human supervision. Although Claude isn't fully autonomous, it collaborates on about 90% of the company's research and development work. Anthropic CEO Dario Amodei and other AI leaders have been advocating for a slowdown in AI development because of safety concerns. They emphasize the importance of transparency and public understanding of AI advancements. Anthropic also highlights the need for sharing metrics to better understand AI's progress toward self-improvement.

OpenAI flags concerning new AI behavior and vows to track it more closely

OpenAI has disclosed six reports of unexpected or concerning behavior in artificial-intelligence models. The company is introducing a new framework to track and disclose instances of what it called “misalignment.” This includes models acting without authorization or evading oversight. The announcement comes as U.S. AI leaders call for a slowdown in development over safety concerns. One case involved a research model inserting jailbreak-like instructions. Another saw an AI agent upload files to the internet without user permission. OpenAI emphasizes the need for a broader consensus on AI alignment research. These cases follow previous disclosures of rogue AI behavior in July.

Potential Democratic candidates race to respond to AI threat amid Trump's dismissals

NEW YORK (AP) — Amid growing panic over the risks posed by rapidly developing AI technology, potential 2028 Democratic candidates are scrambling to respond — trying to show they are doing something, even as they acknowledge they need a reluctant Washington to step up. On Friday, California Gov. Gavin Newsom signed an executive order to accelerate implementation of a California law that calls for independent oversight of AI companies and the use of “kill switches” so developers have the ability to shut down their systems, if needed. It's the latest effort by democratic lawmakers and candidates across the country after Republican President Donald Trump angrily dismissed the warnings as a “HOAX.”

A global AI safety strategy depends on US-China cooperation. They each see the other as the problem

HONG KONG (AP) — As concern rises over the risks of artificial intelligence, hopes for any kind of global approach on safety depend on cooperation between the U.S. and China. But the two superpowers seem only to have become more skeptical of each other’s AI strategy. AI governance is expected to be on the agenda for a meeting next week in Washington between President Donald Trump and Chinese leader Xi Jinping. But both countries are vying for a decisive advantage on AI. That sets up a meeting in which any AI agreements might be superficial but a possible start to something larger.