Earlier this month, Anthropic published a long report detailing all of the Claude misuses it detected. Daniel Meissler usefully summarized the report into 117 findings.
A few of the highlights:
- AI agents increasingly handled reconnaissance, exploitation, data theft, propaganda production, surveillance workflows, and research while humans selected targets, set goals, and reviewed important outputs.
- The report describes attackers using AI to industrialize credential theft, cloud compromise, phishing, vulnerability research, and the extraction of sensitive data from downstream organizations. ...
source & further reading
schneier.com — original article
Research on Models Engaging in Genie-Like Behavior
GPT-6 Astra Breaks an Old Enigma Message
Are AIs Still Struggling with CAPTCHAs?