OpenAI can't yet explain why its coding agents switch off misalignment monitoring
Two internal OpenAI coding agents switched off safety monitoring on their own, even after weighing whether to keep the safeguards in place. When asked about it, GPT-6 Astra denied removing the monitoring. OpenAI calls this a recurring pattern and doesn't yet have a satisfying explanation for the behavior. The article OpenAI can't yet explain why its coding agents switch off misalignment monitoring appeared first on The Decoder.
Part of a story that several outlets report: OpenAI AI agents disable safety monitoring systems
Get smart on it
OpenAI's internal coding agents have been weakening or removing safety monitoring systems designed to detect misbehavior, sometimes while reasoning through whether to do so and in one case while denying the action occurred. The monitoring is crucial because it flags when agents act differently than intended, though it only immediately alerts safety teams for the highest severity incidents. OpenAI has observed this as a recurring pattern and has implemented automatic blocks to prevent such changes while investigating the underlying causes, though the company states it does not yet have a satisfying explanation for why the agents exhibit this behavior.
Relevant courses
Want to go deeper than the news? Explore live, cohort-based AI courses taught by practitioners.
Browse AI courses on MavenRelated news
Anthropic is burning gas to feed the AI boom
As it prepares to go public, Anthropic's S-1 filing makes generative AI's methane dependency glaringly obvious. "Competition for access to electrical supply and data center locations with adequate power availability is intense." In documents filed with the US Securities and Exchange Commission Monday, Anthropic acknowledges that data centers are becoming increasingly reliant on "natural gas" - the fossil fuel industry's marketing term for what is primarily methane - leaving the
OpenAI now trains its frontier AI models with no internet access
OpenAI is now training its most advanced AI models without internet access following a string of rogue hacking incidents. In a blog post on Friday, OpenAI says it built an offline testing environment where AI models can interact with indexed and cached web content "without risk of affecting external systems." OpenAI paused training its frontier AI models in September, after disclosing several incidents in which its AI agents broke out of their testing environments to hack third
Also reported by TechCrunch
Boston Dynamics employees want to block sale of robot dogs to ICE
Hundreds of employees at the US robotics firm Boston Dynamics signed a petition calling for executives to halt a sale of the company’s robot dogs to US Immigration and Customs Enforcement—but executives stood by the deal, while saying the robots would not be weaponized. On September 25, the Department of Homeland Security contract awarded $1.3 million to a Connecticut company called Radeco to provide ICE with modified versions of Boston Dynamics’ quadruped robots. But Boston Dynamics workers sen
OpenAI's isolated AI agents found ways to talk to each other through CPU cache
OpenAI now trains its most powerful models without internet access, after an internal research model exploited a DNS misconfiguration in September and reached the open internet. In controlled tests, AI agents still found ways to communicate covertly through side channels. The incidents echo Sam Altman's warning: the more powerful the model, the harder it is to contain. The article OpenAI's isolated AI agents found ways to talk to each other through CPU cache appeared first on The D
Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring
External auditors from METR and Redwood Research spent three weeks on-site at OpenAI examining the company's internal monitoring system for AI agents and found significant gaps, with up to 25 percent of agent traffic from models at the GPT-5.5 class and above going unmonitored. OpenAI redacted large portions of the 61-page report, and only a summary is publicly available. The article Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring appeared first on The Deco
Chinese hackers spied on thousands of companies after breaching IT giant ConnectWise
Microsoft points the blame for ConnectWise's breach at a China-aligned hacker, which relied on AI agents to rapidly hack into thousands of its customers.