← Back

Wire · regulatory

Lawmakers Push AI Kill Switch Bill Following OpenAI Security Breach

Published

24 July 2026

Topic

regulatory

Sectors

AI Frontier Models

Geography

United States

Source

Read at hothardware.com

Verified

Fusion42 · 24 July 2026 · Fusion42 review

Lawmakers introduced the AI Kill Switch Act following OpenAI's breach, requiring developers of frontier AI systems (trained on $100m+ compute, $500m+ annual revenue) to maintain technical ability to shut down inference, terminate access, or disable systems entirely. The bill grants Homeland Security emergency authority to order shutdowns with $20m/day penalties for non-compliance, though the OpenAI incident that triggered it would fall outside the law's coverage since it occurred during structured testing.

This Wire brief sits within Fusion42's coverage of AI Frontier Models, and 6 sources have reported it between 23 Jul 2026 and 7 Aug 2026.

◆ The Wire takeaway

If you're building frontier AI with $100m+ training compute and $500m+ annual revenue, you now need an always-ready off switch and a 15-day incident reporting clock to Homeland Security, or face $20m daily fines. The threshold is precise enough to catch most frontier labs but loose enough that it will catch you if you scale.

Coverage

6 sources · first reported 23 Jul 2026 · latest 7 Aug 2026

Related on Wire

Topics

AI Frontier Modelsai-regulationemergency-powersfrontier-modelskill-switchcompliance-burden