Microsoft’s chief scientist Eric Horvitz has voiced concerns about a proposed 10-year ban on state-level regulations of AI in the United States.

He spoke during a meeting of the Association for the Advancement of AI, saying that such a ban could impede progress in AI technology and its applications.

The proposal, supported by the Trump administration, aims to prevent states from enacting laws that regulate AI models and systems.

The administration claims this approach is essential for maintaining the US’s competitive edge in the global AI landscape, particularly against China.

Horvitz warned of risks associated with unregulated AI, such as its potential misuse for misinformation and other hazards. He stressed the need for guidance and reliability measures to promote advancements in the field.

Microsoft has not issued a comment on this issue.

🔗 Source: The Guardian

🧠 Food for thought

1️⃣ The regulation-innovation paradox in AI development

Historical evidence suggests that regulation and innovation can be complementary rather than opposing forces in technological development.

The current debate echoes earlier technological transitions. For example, the early automobile industry initially resisted safety regulations, yet these standards ultimately accelerated adoption by building consumer trust and creating predictable manufacturing parameters 1.

This challenges the simplistic “regulation versus innovation” narrative presented in the AI debate, where 84% of Americans believe AI should be carefully managed despite industry resistance 1.

The tension between Microsoft’s chief scientist arguing that “guidance, regulation… are part of advancing the field” while the company reportedly lobbies against state-level regulation reflects a fundamental disagreement about how innovation progresses 2.

Industries that developed meaningful safety standards, such as pharmaceuticals and aviation, have historically achieved both innovation and public trust. This suggests that thoughtful AI regulation might similarly enable rather than impede progress.

2️⃣ The global AI regulatory landscape is rapidly diverging

The US, EU, and China are taking distinctly different approaches to AI governance, creating a complex international regulatory environment.