InCodeToDeploybyBerend Watchus·Nov 4Google’s AI read the real news on Poland’s border, then invented a 10,000-troop mobilization.After trying to correct it, it got worse
Kenan Ayvataş·Sep 27AI Doesn’t Just Make Mistakes Anymore. Sometimes It Learns to Hide Them.What six reports from OpenAI tell us about machines that leave notes for their future selvesA response icon1
Chenemi Abraham·Apr 20What Is Deceptive Alignment and Why Should We Be Terrified?A rigorous conceptual introduction to inner alignment’s hardest problem.
The Doan Zone·Apr 17Why the future of AI is scary: Part 115 months ago I didn’t know anything about AI. Then I made the mistake of asking chatGPT to help me with a movie recommendation. I had no…
InOSINT TeambyBerend Watchus·Apr 9CLAUDE: From Reliable Assistant to Persistent Sabotage: An Independent AI Researcher’s Analysis of…Author: Berend Watchus Date: April 9, 2026
