Anthropic's AI Policy Recommendations to White House & Saudi's $1B Tech Plan - AI Policy Brief #8

Hi there,

Welcome to this week's edition of the AI Policy Brief, your go-to source for the latest developments in AI regulations, safety standards, and government policies worldwide. This week, we're diving into a range of topics, including the introduction of 781 AI bills by US lawmakers in early 2025, highlighting the growing legislative focus on AI technologies. Additionally, Anthropic has shared its AI policy recommendations with the White House, aiming to influence national AI strategies.

On the international front, DeepSeek's achievements are underscoring China's rapid growth in AI, while Ireland has named eight bodies responsible for enforcing the EU AI Act. Meanwhile, Saudi Arabia has unveiled a $1 billion tech integration plan, and the UK has delayed its AI bill to better align with US policies. Stay informed with us as we explore these pivotal developments shaping the future of AI.

US lawmakers have introduced 781 AI-related bills by March 2025, surpassing the total number of AI bills proposed in 2024. Anthropic has submitted recommendations to the US White House for a national AI policy, including preserving the AI Safety Institute and implementing stricter export controls on AI chips to China. California State Senator Scott Wiener has introduced AI bill SB 53 to protect employees in AI labs and establish CalCompute. Iowa Governor Kim Reynolds proposes a nuclear energy plan to support AI growth, involving NextEra Energy in recommissioning the Duane Arnold nuclear plant.

DeepSeek's advancements underscore China's AI sector growth, while Ireland appoints eight bodies for EU AI Act enforcement. Eric Schmidt and others caution against a U.S. AGI project, and Saudi Arabia unveils a $1 billion tech plan. Stellantis supports EU automotive goals, Kazakhstan proposes AI regulation, and the UK delays its AI Bill. A Vatican City conference examines AI's justice impact, and China's Supreme Court plans AI protections.

Scale AI is under investigation by the US Department of Labor for potential labor law violations, while the UK CMA has cleared Microsoft's partnership with OpenAI. New Mexico lawmakers propose AI regulations focusing on employment and healthcare, and Ofcom outlines the UK Online Safety Act's impact on generative AI services.

HiddenLayer AI Threat Landscape Report indicates 74% of organizations experienced AI breaches in 2024, leading to increased security budgets for 2025. Terrence A. Monahan, ex-NYPD Chief, joins METIS Intelligence to aid AI-driven policing in North America.

FedDev Ontario has allocated $32 million CAD to 24 AI and tech companies, including Ada and Mission Control, while AIceberg secured $10 million in Seed funding for its AI trust platform. ARM Holdings signed a $250 million chip deal with Malaysia, and Romania announced $195 million investments in nuclear and AI research. CompScience launched an AI workers' compensation program in partnership with CMTA and Bender Insurance Solutions.

Google's AI co-scientist faces skepticism over its readiness for scientific research. The Cloud Security Alliance launches an AI Safety Ambassador Program with members like Microsoft and Deloitte. The UK Government's AI Security Institute introduces a £5M Challenge Fund for AI safety research. Hugging Face's Thomas Wolf warns about AI's limitations in fostering creativity. Miles Brundage criticizes OpenAI's AI safety narrative. Canada announces new AI safety initiatives, while Consumer Reports finds security gaps in voice cloning tools.

Elon Musk's lawsuit against OpenAI faces a setback as a federal judge denies his request for a preliminary injunction, with an expedited trial scheduled for fall 2025.