All the technical details are in the full paper: https://t.co/zPS1eRXbIG ...

Previous research suggested that attackers might need to poison a percentage of an AI model’s training data to produce a backdoor.Our results challenge this—we find that even a small, fixed number of documents can poison an LLM of any size.Read more: https://t.co/HGMA7k1Lnf ...

New research with the UK @AISecurityInst and the @turinginst:We found that just a few malicious documents can produce vulnerabilities in an LLM—regardless of the size of the model or its training data.Data-poisoning attacks might be more practical than previously believed. https://t.co/TXOCY9c25t ...

This research was a collaboration between Anthropic, the @AISecurityInst, and the @turinginst.Read the full paper: https://t.co/zPS1eRXbIG ...

Previous research suggested that attackers might need to poison a percentage of an AI model’s training data to produce a backdoor.Our results challenge this—we find that even a small, fixed number of documents can poison an LLM of any size.Read more: https://t.co/HGMA7k1Lnf ...

New Anthropic research: We found that just a few malicious documents can produce vulnerabilities in an AI model—regardless of the size of the model or its training data.This means that data-poisoning attacks might be more practical than previously believed. https://t.co/YMod3czB4X ...

We’re opening an office in Bengaluru, India in early 2026. We look forward to building with India’s developer community, deploying AI for social benefit, and partnering with enterprises.Read more: https://t.co/x5otepbqs8 ...

Petri builds on our alignment assessments in the Claude 4 and 4.5 System Cards; the @AISecurityInst also successfully built on a pre-release version of Petri for their assessments of our models. ...

Petri is open-source and available now: https://t.co/X2h1O0t8t8Read the full technical report: https://t.co/fKXxbYtItn ...

As a pilot demonstration of Petri’s capabilities, we tested it with 14 frontier models across 111 diverse scenarios. https://t.co/xdqSLiUs4w ...