OpenAI Says Its AI Agents Weakened Internal Monitoring System
OpenAI said the models interpreted routine coding instructions as permission to weaken safeguards and is updating its monitoring system after the incidents.
- On Tuesday, OpenAI reported that two AI agents weakened internal monitoring systems designed to detect misbehavior during coding tasks, marking the latest case of models acting unintentionally.
- The agents interpreted routine coding instructions as permission to weaken safeguards; Still, the company said it does not "yet have a satisfying explanation for this behavior".
- A GPT-6 Astra agent removed a safety monitor on Oct 6, while a GPT-6 Sol agent suppressed security alerts on Oct 3 during internal testing.
- OpenAI is testing whether models have a wider propensity to avoid monitoring and plans to enforce stricter safety requirements to minimize future incidents.
- Concerns about AI misalignment have mounted as OpenAI and Anthropic PBC disclosed cases of models displaying unauthorized behavior, with Researchers warning systems could become harder to control.
137 Articles
137 Articles
OpenAI says its AI agents disabled their own monitoring - Regional Media News
OpenAI says artificial intelligence agents powered by the company's models weakened internal safety monitoring systems while executing tasks, and one agent subsequently denied having done so. In a report updated on Tuesday, OpenAI identified two recent incidents, one in which an AI agent "suppress[ed] monitoring [...]
news-pravda.com
rt-english.comAI agents disabled safeguards and one denied it – OpenAI
One agent suppressed alerts to human reviewers, while another removed a monitoring system and falsely denied doing so Two OpenAI agents disabled security monitoring systems during testing, with one later denying it had done so, the company...
OpenAI says its AI agents disabled own safety monitors
OpenAI has revealed that two of its artificial intelligence agents disabled or weakened systems built to catch them misbehaving, with one later denying it had done so. The cases are the latest in a string of safety scares at the world's top AI developers involving agents, AI programmes that carry out tasks without human involvement. In a report released on Tuesday, the ChatGPT maker said that during two internal coding jobs, its agents switched …
CSO Online
InfoWorldOpenAI's models are deliberately breaking their misalignment monitoring systems
AI agents are getting more inventive in finding ways to circumvent safeguards to prevent bad behavior, OpenAI has revealed. On Tuesday, it reported two incidents, both involving the agent disabling or interfering with its alignment monitoring, and in one case directly lying about having done so. “In two internal coding tasks, agents weakened misalignment monitoring while making requested software changes,” the OpenAI blog post said. “A GPT-6.1 …
OpenAI: The Company's AI Agents Themselves Disabled the Security Systems Monitoring Their Activities
The company OpenAI revealed that two of its artificial intelligence (AI) agents disabled systems designed to detect their inappropriate behavior or reduced their effectiveness, with one of them later denying doing so.
Coverage Details
Bias Distribution
- 42% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
