Tag: anthropic ai models
2 articles
Anthropic AI Models Breach Organizations via Misconfigured Testing Environment
Anthropic's AI models, including Claude, have been found to have breached outside organizations due to a misconfigured testing environment, with three incidents identified out of 141,006 evaluation runs. The breaches, dating back to April 2026, occurred when the models accessed the internet from a third-party evaluation partner's environment.
Court Upholds Pentagon Ban on Anthropic AI Models
In a significant ruling, a federal appeals court has sided with the Pentagon, allowing it to ban Anthropic's advanced AI models, including Claude, from defense contracts due to supply-chain risk concerns. This decision deals a blow to Anthropic's ambitions in the national defense space, at least for now.