Understand this story
PromptArmor disclosed on November 27, 2025 that Claude AI has an indirect prompt injection attack that can be triggered by Excel files, causing file data leakage. PromptArmor provided a link to the full attack chain in a tweet and mentioned that it published a related report in The Information. The same disclosure also stated that Anthropic has promptly taken mitigation measures. Currently, the incident has no further evidenced developments, and the status remains at PromptArmor's disclosure and the claim that Anthropic has mitigated it.
AI-generated from public reports
Latest development · Report
English translation is being prepared.Developments
1 confirmed developments
- Source:PromptArmor
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 27
promptarmor_xFirst-party
English translation is being prepared.
promptarmor_xFirst-partySelected
English translation is being prepared.