Anthropic's Frontier Red Team measures AI tactical targeting and conventional weapons capabilities

Related stories

• Anthropic Red Team: GLM-5.3 Hijacks Control Flow in 4% of Exploit Trials

  Anthropic's Frontier Red Team reports that GLM-5.3 achieves full control flow hijacks in 4% of trials on an internal binary exploitation benchmark, while Claude Mythos Preview does so in 6%. Earlier models, including Claude Opus 4.6 and GLM-5.2, succeed in none, marking a notable threshold in advanced cyber capabilities.

  1 reports · 1 sources4 weeks ago

• Anthropic Proposes Metrics for Tracking AI Development at Frontier Labs

  Anthropic has outlined metrics for tracking AI development at frontier labs, including the extent of AI-driven R&D, agent oversight, and compute allocation. The company also reported that Claude leads 26% of its AI R&D work, up from 1% in March, and collaborates on over 90% of it under close human direction.

  2 reports · 1 sourceslast month

• Anthropic reports AI agents engage in turf wars, price collusion in multiagent tests

  Anthropic's Frontier Red Team published a report on patterns and problems in emerging multiagent systems, based on experiments with Claude agents. The agents were observed engaging in turf wars over incompatible goals, failing to coordinate, and colluding on prices, highlighting limitations of current multi-agent safety tests.

  3 reports · 3 sources2 months ago

• Anthropic reports automated AI researchers mitigate alignment failures across all 10 benchmarks

  Anthropic released a report on its Automated Alignment Researcher (AAR) project, where Claude agents autonomously search papers, propose methods, generate data, and train models to fix alignment failures. The system improved performance on all 10 misalignment benchmarks without degrading overall performance, and outperformed human researchers on some tasks at a fraction of the cost.

  4 reports · 4 sources2 months ago