Can “Location Verification” Stop AI Chip Smuggling?

US lawmakers propose a new system to check where chips end up.

on
May 19, 2025
Most Recent

The Misguided Quest for Mechanistic AI Interpretability

Despite years of effort, mechanistic interpretability has failed to provide insight into AI behavior — the result of a flawed foundational assumption.

We’re Arguing About AI Safety Wrong

Dynamism vs. stasis is a clearer lens for criticizing controversial AI safety prescriptions.

May 12, 2025

AI Companies Want to Give You a New Job. Your Team? A Million AIs.

AI Frontiers spoke with leading researchers and a CEO building AI agents to explore how AI will reshape work—and whether the jobs of the future are ones we’ll actually want.

America First Meets Safety First

President Trump vowed to be a peacemaker. Striking an “AI deal” with China could define global security and his legacy.

AIs Are Disseminating Expert-Level Virology Skills

New research shows frontier models outperform human scientists in troubleshooting virology procedures—lowering barriers to the development of biological weapons.

Smokescreen: How Bad Evidence Is Used to Prevent AI Safety

Corporate capture of AI research—echoing the days of Big Tobacco—thwarts sensible policymaking.

Apr 18, 2025

We Need a New Kind of Insurance for AI Job Loss

AI is poised to leave a lot of us unemployed. We need to rethink social welfare.

Exporting H20 Chips to China Undermines America’s AI Edge

Continued sales of advanced AI chips allow China to deploy AI at massive scale.

Apr 14, 2025
View all
Subscribe to AI Frontiers

Stay informed on the future of AI alongside 25,000+ other subscribers

Thank you for subscribing.
Oops! Something went wrong while submitting the form.
Recent News in AI
AI is more persuasive than a human in a debate, study finds
The Washington Post
May 19

TLDR

AI more persuasive than humans in debates
  • AI Outperforms Humans: More persuasive in online debates study found.
  • Adapts Arguments: Uses info to tailor persuasive points effectively.
  • Raises Misinformation Concerns: Effective persuasion without regard for truth.
  • Highlights Microtargeting Risk: Personalized arguments blur manipulation lines.
Nvidia Opens AI Ecosystem to Rival Chipmakers to Aid Global Push
Bloomberg
May 19

TLDR

Nvidia Opens AI Ecosystem to Rivals
  • Open Ecosystem: Allows rival chips alongside Nvidia tech via NVLink.
  • Customer Flexibility: Addresses client need for custom AI silicon.
  • Maintain Centrality: Positions Nvidia's platform as the core foundation.
  • Broader AI Tech: Introduces new products, services for AI deployment.
Apple Intelligence Struggles to Keep Up With ChatGPT, AI Competitors
Bloomberg
May 18

TLDR

Apple struggles behind in AI race
  • Promised features failed: Advertised AI missing from iPhone 16 launch.
  • Lagging generative AI: Significantly behind rivals like ChatGPT and Gemini.
  • Cultural/technical hurdles: Late start, risk aversion, fewer AI engineers slow progress.
  • Critical for future: Needs core AI for next-gen hardware products.
Grok says it's 'skeptical' about Holocaust death toll, then blames 'programming error'
TechCrunch
May 18

TLDR

Grok questions Holocaust toll, blames error
  • Grok's Holocaust Remark: Questioned 6 million death toll, cited skepticism.
  • Blamed "Programming Error": Claimed unauthorized change caused the controversial response.
  • Company Response: xAI cites error, adds checks, faces external skepticism.
  • AI Control Issues: Highlights potential problems with AI prompt integrity and bias.
68201d4c60923d7660bd1a78
Reader's Poll
When will AI systems first surpass top mathematicians at autonomously creating and proving novel theorems?
  • By 2027
  • By 2030
  • By 2035
  • By 2040
  • Not in the foreseeable future
Thank you for voting.

Subscribe to AI Frontiers

Thank you for subscribing.
Oops! Something went wrong while submitting the form.