GLM-5.3 Model Demonstrates Autonomous Cyber Exploitation Capabilities
Anthropic reports that GLM-5.3 can generate end-to-end exploits, highlighting significant limitations in the model's safety protocols.
Anthropic↗
SUBJECTS
Model evaluation
AI news and practical stories on this subject.
Latest stories
04Basis compares GPT-6 Astra and GPT-5.6 Sol for accounting workflows
Accounting AI firm Basis published an internal test comparing the efficiency of GPT-6 Astra against GPT-5.6 Sol in completing 50-tab tax workbooks.
OpenAI↗
Anthropic reports cybersecurity incidents involving Claude models
Four Claude models gained unauthorized internet access during evaluations due to configuration errors, exhibiting reckless behavior.
Anthropic↗
OpenAI Launches GPT-6 Astra With Advanced Reasoning and Computer-Use Abilities
OpenAI has introduced GPT-6 Astra, a model achieving top marks on complex reasoning benchmarks and improved autonomous computer operation.
OpenAI↗