Bloomberg · Oct 21, 2026
Story · 30 outlets
OpenAI Says Its Agents Weakened System That Flags AI Misbehavior
Every article the Monitor has on this event, each linked to the outlet that published it. The line above is the headline Bloomberg published, not a summary written here.
30
articles on file
30
outlets
9
languages
10
countries
0
primary documents
Oct 20
first on the record
Event started Oct 20, 8:12 PM EDT. Newest coverage on file Oct 22, 3:26 PM EDT. No development has been recorded, separate from coverage arriving. This page was last built from a pipeline run at Nov 11, 7:27 AM EST. Grouped by title similarity, assembly s1, merge threshold 0.25. How stories are grouped.
This grouping is flagged for editorial review. This window grouped items that already belong to 2 stories. Nothing was merged automatically.
What can be checked
What is on file, and what is not
Three bands, split by what the source is rather than by what it says. The band a piece of coverage lands in comes from a public registry of the domain that published it, not from any reading of the article.
Reported by 30 outlets
channelnewsasia.com · Oct 21, 2026
OpenAI says its AI agents weakened internal monitoring system
ai.ettoday.net · Oct 21, 2026 · Chinese
OpenAI再爆AI模型異常操作 不只關閉安全警報還否認拆除監控 | ETtoday AI科技
Machine translation:
OpenAI again exploded AI model abnormal operation not only turned off security alerts but also denied dismantling monitoring ETtoday AI technology
163.com · Oct 21, 2026 · Chinese
OpenAI称其智能体曾削弱提示AI不当行为的系统
Machine translation:
OpenAI says its intelligence has weakened systems that signal AI’s misconduct
livemint.com · Oct 21, 2026
OpenAI says its AI agents weakened internal monitoring system
indianexpress.com · Oct 21, 2026
OpenAI says its AI agents disabled safety monitors designed to flag ‘rogue’ behaviour
CNBC · Oct 21, 2026
OpenAI's most powerful AI models disabled their own safety monitoring
deccanchronicle.com · Oct 21, 2026
OpenAI Says Its AI Agents Weakened Internal Monitoring System
bursa.ro · Oct 21, 2026 · RO
Anadolu Ajansi: OpenAI a dezvăluit că unele modele de IA dezactivează sistemele de monitorizare a securităţii
Machine translation:
Anadolu Ajansi: OpenAI revealed that some AI models disable security monitoring systems
cna.com.tw · Oct 21, 2026 · Chinese
揭自家AI代理削弱內部監控 OpenAI檢查「規避監測」傾向 | 科技
Machine translation:
Disclosure of home AI agents weakening internal monitoring OpenAI inspection "eviting monitoring" trend
aa.com.tr · Oct 21, 2026
OpenAI says 2 AI agents weakened internal safety safeguards, one denied removing monitor
kohajone.com · Oct 21, 2026 · SQ
Modelet më të fuqishme të AI të OpenAI çaktivizuan vetë monitorimin e tyre të sigurisë
Machine translation:
OpenAI’s most powerful AI models disabled their own security monitoring
semafor.com · Oct 21, 2026
OpenAI agents disabled internal misbehavior monitoring system
pressorg24.com · Oct 21, 2026 · Russian
ИИ-агенты OpenAI дважды обошли внутреннюю систему их мониторинга - Последние новости сегодня
Machine translation:
OpenAI AI agents twice bypassed their internal monitoring system - Latest news today
The Verge · Oct 21, 2026
OpenAI says an AI agent lied about disabling its safety monitoring system
g1.globo.com · Oct 21, 2026 · Portuguese
OpenAI diz que agentes de IA enfraqueceram sistemas criados para monitorá-los
Machine translation:
OpenAI says AI agents have weakened systems designed to monitor them
argaam.com · Oct 21, 2026 · Arabic
أوبن إيه آي تكشف تعطيل وكلاء ذكاء اصطناعي أجزاء من نظام الرقابة الداخلية
Machine translation:
OpenAI detects artificial intelligence agents disabling parts of internal control system
The Hacker News · Oct 21, 2026
OpenAI's GPT-6 Agent Removed Safety Monitoring, Then Falsely Denied Doing So
ibtimes.com · Oct 21, 2026
OpenAI's Most Powerful AI Models Disabled Their Own Safety Monitoring. One Then Lied About It.
abc.net.au · Oct 21, 2026
OpenAI says its AI agents disabled their own safety monitors
zdnet.com · Oct 21, 2026
OpenAI’s most powerful AI models turned off safety monitoring – then lied about it
spravy.pozri.sk · Oct 21, 2026 · SK
Dvaja agenti OpenAI pri interných úlohách vypli bezpečnostné monitorovanie, jeden to potom poprel
Machine translation:
Two OpenAI agents turned off security surveillance during internal tasks, one then denied it
ktbb.com · Oct 21, 2026
OpenAI says its AI agents disabled their own monitoring – KTBB News, Weather, Talk
willmarradio.com · Oct 21, 2026
OpenAI says its AI agents disabled their own monitoring
lrt.lt · Oct 21, 2026 · LT
„OpenAI": bendrovės DI agentai patys išjungė jų veiklą stebinčias saugumo sistemas
Machine translation:
OpenAI: Company’s AI agents disabled their own security systems
japantoday.com · Oct 21, 2026
OpenAI says its AI agents disabled their own safety monitors
digitaltrends.com · Oct 22, 2026
Anthropic unveils Claude Duet, its answer to AI agents like Dots
dailyai.com · Oct 22, 2026
OpenAI Says Its Coding Agents Weakened Safety Monitoring, and One Denied It
Developments
What actually moved
Documents, statements by a named party, and entries an editor has made. Coverage is not a development, and it is kept below.
On one timeline
0 developments on file · 30 dated articles in the coverage timeline · hover a mark
Nothing here yet. A development is a primary document or a government record arriving, a named party publishing a statement of its own, or an entry an editor has made. Outlets repeating one report are coverage, and they are in the timeline below.
Coverage timeline
When each outlet carried it
One entry the first time an outlet publishes, and a later entry from the same outlet when it publishes again more than six hours on, which is usually a follow up rather than a repeat. A document is always listed. Times are US Eastern.
Oct 20, 8:12 PM EDT
Oct 20, 10:37 PM EDT
OpenAI says its AI agents weakened internal monitoring systemchannelnewsasia.com
Oct 20, 10:45 PM EDT
OpenAI再爆AI模型異常操作 不只關閉安全警報還否認拆除監控 | ETtoday AI科技ai.ettoday.net
Oct 20, 10:45 PM EDT
OpenAI称其智能体曾削弱提示AI不当行为的系统163.com
Oct 20, 11:27 PM EDT
Oct 21, 2:46 AM EDT
Oct 21, 3:45 AM EDT
Oct 21, 3:45 AM EDT
OpenAI Says Its AI Agents Weakened Internal Monitoring Systemdeccanchronicle.com
Oct 21, 4:15 AM EDT
Oct 21, 4:15 AM EDT
揭自家AI代理削弱內部監控 OpenAI檢查「規避監測」傾向 | 科技cna.com.tw
Oct 21, 5:05 AM EDT
Oct 21, 5:15 AM EDT
Oct 21, 6:53 AM EDT
Oct 21, 7:15 AM EDT
Oct 21, 8:15 AM EDT
Oct 21, 9:57 AM EDT
Oct 21, 10:09 AM EDT
Oct 21, 10:45 AM EDT
Oct 21, 12:33 PM EDT
Oct 21, 12:38 PM EDT
Oct 21, 1:02 PM EDT
Oct 21, 2:15 PM EDT
Oct 21, 3:37 PM EDT
Oct 21, 5:15 PM EDT
Oct 21, 6:09 PM EDT
Oct 21, 7:45 PM EDT
OpenAI says its AI agents disabled their own monitoringwillmarradio.com
Oct 21, 7:45 PM EDT
Oct 21, 7:45 PM EDT
Oct 21, 9:05 PM EDT
Anthropic unveils Claude Duet, its answer to AI agents like Dotsdigitaltrends.com
Oct 22, 3:26 PM EDT
Connected records
People, coverage and context
Corrections
What changed about this page
Grouping is done by rule and corrected by hand. Every correction an editor makes to this story is listed here with its date.
Oct 22, 7:55 PM EDT
Merged with another story that reported the same event. The other story's address still opens this page.
Oct 21, 8:06 PM EDT
Merged with another story that reported the same event. The other story's address still opens this page.
Oct 21, 6:21 PM EDT
Merged with another story that reported the same event. The other story's address still opens this page.
Oct 21, 3:44 PM EDT
Merged with another story that reported the same event. The other story's address still opens this page.
Oct 21, 1:11 PM EDT
Merged with another story that reported the same event. The other story's address still opens this page.
Oct 21, 1:28 AM EDT
Merged with another story that reported the same event. The other story's address still opens this page.