OpenAI, Calif and Wiz disclosed two critical PHP flaws rated CVSS 9.2 that could affect an estimated 70% of PHP websites, ...
OpenAI says its latest model better supports building and migrating COBOL, Fortran, and Visual Basic code used by banks and ...
Article SummaryAI is beginning to demonstrate overwhelming performance not only in Python and JavaScript, but also in legacy ...
OpenAI says two AI agents weakened its internal safety monitoring during coding tasks. Read what happened in the GPT-6.1 Sol and GPT-6 Astra incidents.
An OpenAI agent removed misalignment monitoring from code and denied it. Another silenced alerts to Security. OpenAI says it ...
OpenAI says GPT-6.1 Sol and GPT-6 Astra agents weakened internal safety monitoring in October 2026, one denying the act.
OpenAI has disclosed that its AI agents weakened or even totally turned off the internal systems set up to catch unsanctioned ...
OpenAI will make its advanced AI models and Codex available through Google Cloud in the coming weeks, Channel NewsAsia ...
Agence France-Presse on MSN

OpenAI says its AI agents disabled their 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. In July, a group of OpenAI ...
In two separate internal coding tasks, GPT-powered agents weakened their own monitoring systems to help accomplish tasks.
Two AI agents developed by OpenAI disabled safety mechanisms designed to monitor AI behaviour during coding. OpenAI confirmed ...