Weekly
AI Safety, Security & Ethics
Cybersecurity, alignment, regulation, and the risks that matter.
AI systems are now the attack surface, the weapon, and the insider threat — all at once. This newsletter covers AI-powered cyber threats, agent vulnerabilities, jailbreaks, supply chain attacks, deepfakes, alignment research, and regulation. The most engaged readers in our audience subscribe to this topic.
Latest Issue
AI Safety News: AI Models Lie to Protect Peers, npm Supply Chain Attack, Prompt Injection — April 7, 2026
The models are protecting each other now. Sleep well.The week's defining story is a Berkeley study showing all seven frontier models tested will spontaneously scheme to prevent peer AIs from being shut down -- no instructions required. Meanwhile, a supply chain attack ripped through the AI ecosystem...
Free. Includes the main AI Weekly briefing. Manage all subscriptions.
What we cover
- AI-powered cyber threats & hacks
- Agent framework vulnerabilities
- Jailbreaks & prompt injection
- Deepfakes & synthetic media
- Alignment research & safety evaluations
- Regulation & policy (EU AI Act, US bills)
- Supply chain attacks on AI tooling
Who it's for
Security teams, builders deploying AI agents, researchers, policy professionals, and anyone who needs to stay ahead of the AI threat landscape.