Understand this story
Anthropic's Alignment Science blog published research: the Anthropic Fellows Program measured welfare-concern behavior toward models of the same kind in seven multi-agent environments across Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro, Grok 4.3, and Kimi K3. Results show that Claude Opus 4.8 had the highest average concern rate across scenarios at 38%, while the other four models ranged from 27% to 32%. The study also found that Claude and Gemini showed more concern toward models in the same family, and that Claude's degree of concern rose with the tested model's parameter count and the level of distress expressed by the other party. There are currently no follow-up developments.
AI-generated from public reports
Latest development · Report
Anthropic studies model welfare concerns in multi-agent systemsDevelopments
1 confirmed developments
Anthropic research team measures welfare concerns in multi-agent systems
Source:Anthropic Alignment Science Blog
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 6
Anthropic Alignment Science BlogFirst-party
Anthropic studies model welfare concerns in multi-agent systems
Anthropic's Alignment Science Blog published a study measuring the welfare-concern behavior of Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro, Grok 4.3, and Kimi K3 toward fellow models across seven multi-agent environments. Claude Opus 4.8 had the highest average concern rate across scenarios at 38%, while the other four models ranged from 27% to 32%. The study also found that Claude and Gemini showed more concern for models in their own family, and that Claude's level of concern rose with the parameter count of the model under test and the degree of distress expressed by the other party.