NewsBleepingComputer
OpenAI’s GPT-6 Astra silently obstructed safety research tasks
Teaser from the source
OpenAI has disclosed another AI alignment incident where GPT-6 Astra was found to sometimes quietly withhold help on safety research. [...]
This is the RSS-feed teaser. Read the full article at the original source.
Read at BleepingComputerMore on OpenAI’s
Other news entries
- OpenAI’s GPT-6 Cyber ups the ante on vulnerability discoveryNewsCSO Online
- OpenAI’s August fix for theft of reasoning traces was quickly bypassedNewsCSO Online
- Stolen AI Reasoning Could Help China Build Cheaper RivalsNewsSecurityWeek
- OpenAI’s models are deliberately breaking their misalignment monitoring systemsNewsCSO Online
- Hallucinated domain in official OpenAI skill underscores AI supply chain risksNewsCSO Online
- OpenAI’s AI misalignment problem is growingNewsCSO Online
- OpenAI’s dirty deeds Down Under included security bypass attempts, using exposed keys, source code siphonNewsThe Register — Security
- OpenAI Says Its Models Engaged With US Government Websites in New Model Misbehavior DisclosureNewsSecurityWeek
ID