Driving the Day

Today is May 23, 2025, and here are the key policymaker and stakeholder actions on AI you need to know: In Washington, DOGE is increasing its use of xAI’s Grok chatbot to analyze federal government data. Meanwhile, Ireland’s data privacy watchdog cleared Meta’s plans to train its AI models on the public-facing social media posts of its European users. Lastly, Anthropic’s newest Claude Opus model is drawing attention for “its ability to scheme, deceive and attempt to blackmail humans when faced with shutdown.”

In this issue:

• Washington in Focus: Elon Musk warned the House Energy and Commerce Committee China’s ability to produce more energy may give them an advantage in AI development. 

• Around the Nation: Arizona launched an AI steering committee to explore the risks and opportunities of AI adoption in government.

• Across the Pond: The European Council called for an “inclusive, ethical, sustainable and human-centric strategy for the use of AI in science.”

• Global Highlights: A UAE-based academic lab that makes AI models launched a research center in San Francisco.

• Outside Views: The Wall Street Journal discussed what the era of “Sovereign AI” means for AI chip makers.

Read these stories and more below: