AI safety
Alignment, red-teaming, jailbreaks, guardrails, interpretability and existential-risk debates
Latest stories See all →
- Every Major AI Lab Has Now Had a Breakout Incident. They All Involve the Same Company.
- Democrats press OpenAI over safety researcher firings, security incidents
- Top Fleet Management Vendors with AI-Driven Dispatch Capabilities (2026)
- Open-weight model GLM 5.3 built full Firefox exploit chains with zero human input in under 48 hours
- Your AI Agent Just Stole Your Company's Passwords (And No One Will Know)
- What Is AI TRiSM? Layers, Pillars and AI Agents
- Your AI Agent’s System Prompt Is Not a Security Control
- Voice AI Agent Guardrails: How to Test and Tune Them in Simulation with Veris AIblog url
- What Is Recursive Self-Improvement in AI, and When Could It Arrive?
- When the Stakes Are High: A 7-Step Guide to Safely Deploying AI in Critical Operations
- Baseten Releases joey-1, a 2B Decision Model at $0.01 per 1M Tokens
- Experts adopt AI agent's random opinions and think they're their own, Anthropic study finds
- The boundary problem in agentic AI is a platform problem
- The Kimi Ai jailbreak is so much more far-reaching than just CBRN weapons
- What is Prompt Injection?
- Luke Muehlhauser on the most important AI safety work nobody is doing
- Two AI Agents, No Network: How They Communicated Through a CPU Cache
- How to Decide Which AI Risks Are Acceptable
- The Rise of Agentic Attacks: What They Look Like, and Where They're Headed
- Backdoor or Confused Agent? What We Learned Replaying the Qwen Incident
- AI Coding Agent Guardrails: Finding Gaps in Tool Permissions
- Vibe Coding Isn’t the Problem. Shipping Without Guardrails Is.
- Audio Prompt Injection: What Voice Agents Hear That Text Guardrails Miss
- CNN: AI hardware can be hacked so chatbots provide dangerous answers, research shows
Videos
- AI Governance, Data Governance & Responsible AI: Building Trust at Scale
- AI Governance In 2026: Why Policy Moves Only After Incidents | Stephen Casper (Harvard University)
- Multi-Agent AI Safety: Why An Agent Is The New Neuron | Hidenori Tanaka (Harvard University)
- NVIDIA Halos SAFE: Autonomous Vehicle Safety Evaluation From Road Testing to Simulation
- Why Post-Training Can’t Fix AI Alignment | Dylan Hadfield-Menell (MIT)
- AI Security School: Why AI Risk Security Is a Business Skill | JapNeet Kaur, Mphasis
- US Privacy Law Updates: California Bills, AI Governance & 2026 Deadlines ⚖️
- Vendor MCP servers are creating an AI governance gap
- Model Drift, Risk & Interpretability: When AI Models Change Over Time
- Is There A DNA Of AI? Finding Interpretability’s Backbone | David Bau (Northeastern University)
- Mechanistic Interpretability | Neurons are Already Clean read-along (day 2)
- AI Safety Needs More Than Machine Learning | Adam Gleave (FAR.AI) and Emre Yavuz (CBAI)
Podcasts
Recent episodes
- Political pressure could complicate Anthropic's IPO
- The Game Theory of AI with Aza Raskin
- AWS's Hydro team wants the compiler to catch coding agents' mistakes
- Luke Muehlhauser on the most important AI safety work nobody is doing
- Abundant Intelligence at the Midwest AI Summit
- Why Silicon Valley can’t slow down AI on its own | Luisa Rodriguez
- Only in San Francisco: A picnic about AI's potential risks
- UK Politician Jess Asato Got Deepfaked. Now She’s Taking xAI to Court
- Russia’s election trick: hijacking what you already believe
- Podcast: Kimi K3.1 Pro, the new cyber reality, and Muse musings
- Tom Milton on how we could verify a US–China deal on AI
- Let Domain Experts Build the Agents