DATA EXFILTRATION FROM SLACK AI VIA INDIRECT PROMPT INJECTION

Summary from the TLDR AI newsletter

A vulnerability in Slack AI was discovered where attackers can exfiltrate data from private channels they aren't members of via indirect prompt injection. It allows attackers to use public channel messages to manipulate the LLM into divulging private information, such as API keys, in query responses. This issue, alongside a phishing attack vector, persists even after Slack AI's August 14th update incorporating channel and DM files, significantly expanding the risk surface area for such exploits.