Trump's Cybersecurity Shake-up, UNESCO's AI Roadmap, and UK Delays AI Regulation - AI Policy Brief #21

Hi there,

Welcome to this week's edition of the AI Policy Brief, your go-to source for the latest updates on AI regulations, safety standards, and government policies worldwide. This week, we're covering a range of topics from the U.S. to the Caribbean. In the U.S., Trump has revised the national cybersecurity policy, while Anthropic's CEO has voiced opposition to a proposed AI regulation ban. Meanwhile, Sen. Cruz has suggested cutting broadband funds to support AI regulation efforts.

Internationally, UNESCO has unveiled an AI policy roadmap at the Caribbean Media Summit, and the AFIA has warned against overly restrictive AI regulations. In the UK, CERSI-AI has launched to regulate AI in healthcare, and Thailand has introduced its first AI bill aimed at responsible use. Stay tuned for more insights and developments in the world of AI policy.

President Donald Trump has signed an executive order revising the U.S. cybersecurity policy, removing initiatives from former President Joe Biden. Dario Amodei, CEO of Anthropic, opposes a proposed 10-year moratorium on AI regulation, advocating for federal transparency standards. Senator Ted Cruz suggests withholding federal broadband funding from states that regulate AI, as part of a broader bill to ban state AI regulation for 10 years.

UNESCO introduced an AI policy roadmap at the Caribbean Media Summit, while the Australian Finance Industry Association warned against restrictive AI regulations. CERSI-AI launched in the UK to regulate AI in healthcare, and Namibia is developing AI legislation. Thailand introduced its first AI bill for responsible use.

The FTC is addressing deceptive AI claims by enforcing consumer protection laws, with recent actions against Workado and Cleo AI for misleading statements about their products. The Digital Advertising Alliance is reviewing its Self-Regulatory Principles to guide the use of AI in interest-based advertising, aiming for responsible consumer engagement.

OpenAI has dismantled 10 influence operations linked to the Chinese government, while a UAE-US data centre project involving Nvidia and OpenAI faces delays over security concerns. Anthropic appoints Richard Fontaine to its governing trust for AI models in U.S. national security.

The U.S. Food and Drug Administration has launched 'Elsa', a generative AI tool for scientific reviews. The U.S. AI Safety Institute has been renamed to the Center for AI Standards and Innovation by the Trump administration. The UK government has partnered with Google, Microsoft, and Amazon for a £187 million AI skills initiative. The UK's Financial Conduct Authority and Nvidia have developed a 'Supercharged Sandbox' for banks. The South Australian Government announced a $28 million AI program for public services.

Microsoft will implement a ranking system for AI models based on safety performance, aimed at enhancing data protection for its cloud customers. NHS Lothian and NHS Greater Glasgow and Clyde are collaborating with AI evaluation company Aival on a £1m project to assess AI safety in healthcare, focusing on triaging stroke and lung cancer patients. The British Film Institute (BFI) has released a report with nine recommendations aimed at helping the UK screen sector navigate the challenges and opportunities of AI technology.

The UK High Court has warned lawyers about AI misuse in citing non-existent cases. The copyright lawsuit between Getty Images and Stability AI has commenced, with Getty Images alleging unauthorized use of its images. OpenAI is appealing a court order to retain ChatGPT data in a copyright case with The New York Times.