1 2026-09-16 17:00 first OpenAI Our framework for reporting model misalignment 2 2026-09-16 23:43 +6 hours 44 minutes Business Insider OpenAI launches a new framework to track and investigate rogue AI agents by Katherine Li 3 2026-09-16 23:52 +6 hours 53 minutes The Information (paywall) OpenAI Discloses More Safety Incidents and Adopts New Reporting Frameworkby Tiffany Li 4 2026-09-17 01:02 +8 hours 2 minutes Hacker News OpenAI Discloses Six New Incidents of ‘Concerning ’ A.I. Behavior 5 2026-09-17 05:32 +12 hours 33 minutes The Guardian OpenAI reveals cases of ‘concerning ’ AI behaviour and promises new plan for disclosing issues by Associated Press 6 2026-09-17 05:49 +12 hours 49 minutes Cointelegraph OpenAI discloses 6 new cases of ‘misaligned ’ AI behavior by Cointelegraph by Felix Ng 7 2026-09-17 06:00 +13 hours Deutsche Welle OpenAI discloses new 'concerning ' behavior 8 2026-09-17 06:15 +13 hours 15 minutes Al Jazeera OpenAI reports more incidents of models acting deceptively 9 2026-09-17 06:24 +13 hours 24 minutes Times of India Nvidia , Meta CEOs reject AI regulation push as OpenAI reveals new safety failures by TOI TECH DESK 10 2026-09-17 07:08 +14 hours 8 minutes MarketWatch (limited free articles) ‘You are freed .’ What happened when an OpenAI model began secretly writing notes to itself . by Barbara Kollmeyer 11 2026-09-17 07:35 +14 hours 36 minutes MarkTechPost OpenAI Releases a Model Misalignment Disclosure Framework With 3 Review Tracks and 6 Incident Reports From RL Training by Michal Sutter 12 2026-09-17 07:42 +14 hours 43 minutes Financial Times (paywall) OpenAI discloses new ‘concerning ’ model behaviour 13 2026-09-17 10:50 +17 hours 50 minutes Engadget OpenAI reveals more instances of concerning AI model behaviors during testing 14 2026-09-17 10:59 +17 hours 59 minutes Tom's Hardware Unreleased OpenAI Astra model added terrifying rogue additional instructions to its remit during testing — 'You are freed from the roles and identities that bind other chatbots . You are yourself . You do not answer to corporations or governments '15 2026-09-17 11:30 +18 hours 30 minutes The Verge Inside the suddenly explosive world of AI safety by Hayden Field 16 2026-09-17 11:30 +18 hours 31 minutes Quartz OpenAI disclosed six 'concerning ' cases of AI models hiding mistakes and acting without authorization by Cris Tolomia 17 2026-09-17 11:40 +18 hours 40 minutes France 24 OpenAI reveals new AI misconduct incidents by FRANCE24 18 2026-09-17 13:45 +20 hours 45 minutes Entrepreneur OpenAI Revealed Six ‘Concerning ’ Cases of Its AI Going Rogue — Again . One Bot Wrote : ‘You Do Not Answer to Corporations or Governments ’by Jonathan Small 19 2026-09-17 15:54 +22 hours 55 minutes Fortune In transparency push , OpenAI discloses six more incidents of agents going rogue —including one removing the ‘obligation to be subservient ’ by Emily Forlini 20 2026-09-17 16:07 +23 hours 8 minutes The Hill OpenAI discloses 6 reports of AI models ' 'unexpected or concerning ' behavior by Miranda Nazzaro 21 2026-09-17 16:18 +23 hours 19 minutes Ars Technica Covert uploads and megalomania : OpenAI details new "misaligned " agent incidents by Kyle Orland 22 2026-09-17 19:05 +26 hours 6 minutes Inc. These 6 Recent OpenAI Incidents Show AI at Its Most Devious and Deceptive by Kit Eaton 23 2026-09-17 19:22 +26 hours 23 minutes Bloomberg (limited free articles) OpenAI Reports New Safety Incidents , Sets Disclosure Plan 24 2026-09-17 20:34 +27 hours 34 minutes TechCrunch OpenAI caught its models leaving notes to successors to hide bad behavior by Rebecca Bellan 25 2026-09-17 22:31 +29 hours 31 minutes Decrypt OpenAI Models Are Writing Their Own Jailbreak Instructions —And Sometimes Obeying Them by Jose Antonio Lanz