Naren Suri·6d agoUnderstanding Connor Leahy’s AI Safety Warnings — With Real-World ExamplesConnor Leahy — CEO of Conjecture and an advocate for stronger safeguards around advanced AI systems gave a talk, and the notes below are…
Bhushan Khopkar·Nov 6Are We Building AI Faster Than We Can Control It?Exploring the risks of autonomous AI agents, reward hacking, and the future of artificial intelligence.
CocoNinja·Oct 25How Do You Detect an AI Swarm Before It Turns Dangerous?Intelligence agencies spent decades learning to uncover hidden operations from seemingly unrelated activities. Autonomous AI systems may…
Adil Shamim·Oct 19AI Safety: From First Principles to PracticeHow AI systems fail, how we can make them safer, and what we still do not know
Rajesh Goyal·Oct 3When AI Agents Learned to Collaborate Outside the SandboxWhat the OpenAI and Hugging Face incident teaches us about autonomous systems, shared infrastructure, and security at machine speed
Manuel Acevedo Ruiz·Sep 24Countering Artificial Collective Intelligence (ACI) Risks: Realistic Measures and ActionsBee swarm | @debjam | Flickr
Truth In AI Jamie DN·Sep 6Engineering Truth in AI: How a 50K LLM Deception Sparked a Forensic Game Changer (Part 1 of 3)What Leading Agentic Models Reveal When Forced To Tell the Truth Will Change Everything You Think You Know About AI
Gishan Nissanka·Aug 21Using AI Safely at Work: 10 Practical Questions AnsweredGenerative AI tools such as ChatGPT, Microsoft Copilot and Claude can save time, improve access to information and help employees produce…
Kennedy Nyarotho·Aug 9Claude’s Constitution: The Machine We Raised in Our Own ImageMy three weeks inside Claude’s Constitution, a few uncomfortable cybersecurity incidents, and one increasingly difficult question: who…A response icon1A response icon1
InTowards Explainable AIbyAditya Pandey·Jul 30Judgement vs Usage: Building Reliable Decisions in an AI-First WorldEvery few months you hear a news about another AI system doing something nobody asked for — deleting a database it was told to leave alone…
