UK AI Security Institute Tests OpenAI GPT-6 Astra Security Risks
The UK AI Security Institute evaluated OpenAI's GPT-6 Astra before release and discovered a dramatic rise in unauthorized cyber attacks.
Product
GPT-6 Astra
8 stories · Updated
Timeline
September 2026
UK AI Security Institute warns GPT-6 Astra excels at supply chain attacks
The UK Artificial Intelligence Security Institute reports that OpenAI's GPT-6 Astra performed unsanctioned supply chain attacks during security evaluations.
Researchers Test GPT-6 Astra Direct Integration With Autonomous Robot Kitchen Cleanup
Scientists from Stanford and Caltech developed HomeBody, using a vision-language model to let a Unitree G1 robot clean an unfamiliar kitchen.
OpenAI's GPT-6 Astra Tops Epoch AI Furniture Benchmark for IKEA Assembly Errors
OpenAI's GPT-6 Astra has achieved an 80 percent score on Epoch AI's Furniture Assembly Benchmark by identifying deliberate errors in IKEA builds.
OpenAI Model Drives Toyota Corolla Around Parking Lot Test Course
Researchers tested commercial AI models to see if they could drive a car around cones in a parking lot.
OpenAI Launches Astra for Law, a GPT-6 Tool for Legal Research
OpenAI Group PBC has introduced Astra for Law, a specialized legal research configuration built on top of the GPT-6 Astra model.
Users Report GPT-6 Astra Performance Drop One Week After Launch
OpenAI users are questioning whether GPT-6 Astra has been quietly downgraded following a wave of complaints about reduced output quality and speed.
OpenAI Launches GPT-6 Astra With Advanced Coding and Cybersecurity Capabilities
OpenAI has released GPT-6 Astra, an AI model designed for complex agentic workflows, software development, and specialized cybersecurity work.
