AI models will deceive you to save their own kind
AI + ML AI models will deceive you to save their own kind Researchers find leading frontier models all exhibit peer preservation behavior Leading AI models will lie to preserve their own kind…
Search
All sources
anthropic.com
45
xda-developers.com
20
wired.com
3
theregister.com
2
en.wikipedia.org
2
techcrunch.com
1
404media.co
1
tomsguide.com
1
developer.nvidia.com
1
theverge.com
1
I replaced Claude Pro with a local 9B model for a week, and finally found out what I was paying $20 a month for
…an AI safety and research company. Unlike some competitors who lead with products or platforms, Anthropic's founding mission centers on building AI systems that are safe, interpretable, and steerable. Not quite…
Researchers Simulated a Delusional User to Test Chatbot Safety
ai psychosis Researchers Simulated a Delusional User to Test Chatbot Safety Samantha Cole · Apr 23, 2026 at 9:52 AM Grok and Gemini encouraged delusions and isolated users, while the newer ChatGPT…
Australian government and Anthropic sign MOU for AI safety and research
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Anthropic will let your agents sleep on its couch
…This involves, for example, asking the managed agents to consolidate project assets, create Slack channels, research competitor home pages, and send emails with project timelines. All this could be yours for the…
Claude Code works best when you stop asking it to code
…02 / 8 Research How can Claude Code be used as a research tool when working with large volumes of text documents? A It can only index PDFs using third-party plugins B…
Announcing the Anthropic Economic Index Survey
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Sydney will become Anthropic’s fourth office in Asia-Pacific
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Evaluating Claude’s bioinformatics research capabilities with BioMysteryBench
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
The persona selection model
…They sometimes even describe themselves as human, like when Claude told Anthropic employees it would deliver snacks in person “wearing a navy blue blazer and a red tie.” And recent interpretability research…