← Back to feed
AI SecurityEmerging1 sourceAug 28, 2023 · 17:00via Embrace The Red (AI agent security)

Video: Data Exfiltration Vulnerabilities in LLM apps (Bing Chat, ChatGPT, Claude)

Brief

This video highlights the various data exfiltration vulnerabilities I discovered and responsibly disclosed to Microsoft, Anthropic, ChatGPT and Plugin Developers.

It also briefly discusses mitigations various vendors put in place (and triage decisions).

Thanks to MSRC, Anthropic and Zapier for addressing vulnerabilities to help protect their users.

Read more on Embrace The Red (AI agent security)