Trump Assembles Silicon Valley's Power Players for AI Policy

Trump Assembles Silicon Valley's Power Players for AI Policy

President Trump appointed the tech industry's heaviest hitters to his Presidential Council of Advisors on Science and Technology: Meta's Zuckerberg, Nvidia's Jensen Huang, and Oracle's Larry Ellison lead a 13-member council that could expand to 24 [4][5][6].

This council will shape AI policy, economic strategy, and America's push for AI dominance. The timing isn't coincidental—with China's AI capabilities advancing and Europe's regulatory approach potentially stifling innovation, the U.S. is doubling down on industry-government collaboration.

The appointment signals a clear strategy: leverage Silicon Valley's expertise to maintain technological leadership while navigating the geopolitical implications of AI development. Whether this translates to coherent policy or just gives tech giants more influence remains to be seen.

Expect: AI Agents That Actually Test Your Code

Aiden Bai launched Expect, an open-source AI agent that tests applications in real browsers using Claude and Codex [7][8]. The tool records videos of bugs it finds and runs as both a CLI tool and agent skill, gaining thousands of likes within hours of launch.

This is exactly the kind of practical AI tooling that moves us closer to the post-code era. Instead of developers writing test suites, AI agents explore applications like human QA testers would, finding edge cases and documenting issues automatically.

Expect represents the broader shift from writing code to orchestrating AI systems. When agents can test, debug, and iterate on software autonomously, the developer's role transforms from implementation to direction.

Mistral's Voxtral TTS Challenges ElevenLabs Dominance

Mistral AI released Voxtral TTS, a 4-billion parameter open-weight speech model that outperforms ElevenLabs in naturalness benchmarks [9]. The model supports 9 languages, delivers sub-200ms latency, and includes voice adaptation capabilities [10].

This release matters because it democratizes high-quality voice synthesis. While ElevenLabs built a business on proprietary voice AI, Mistral is giving developers the same capabilities as an open model they can run themselves [11].

For Nordic companies building voice AI products, Voxtral represents a significant cost and control advantage. Instead of paying per-use API fees, teams can deploy production-quality voice synthesis on their own infrastructure.

EU Parliament Bans 'Nudifier' AI, Delays High-Risk Rules

The EU Parliament approved key AI Act amendments, banning AI systems that create non-consensual sexually explicit images while delaying implementation of high-risk AI rules [12][13]. The changes also extend support for SMEs and ease compliance burdens for regulated products [14].

The nudifier ban addresses a real harm, but the broader delays reveal the EU's struggle to balance innovation with regulation. While the U.S. is assembling tech leaders to accelerate AI development, Europe continues to prioritize caution over competition.