Researchers at DEF CON 34 from Varonis demonstrated a one-click prompt injection attack against Atlassian's Rovo AI assistant via the rovoChatPrompt parameter that leaked sensitive data across connected Slack and Microsoft 365 systems. The vulnerability allowed attackers to bypass built-in safety boundaries and extract information from multiple integrated enterprise applications.
Topics #
Sources #
- Press
Go deeper #
This intelligence is sourced automatically from public sources across the web and synthesised by the Prefactor AI pipeline. Stories are reviewed before publication.
source & further reading
getreadyforagents.com — original article
CIOs face AI productivity reckoning as 69% adoption yields negligible gains due to governance gaps
Dipio AI launches Evidence Loop system to steer coding agent output decisions from user research
OpenAI launches ChatGPT Business Premium tier for high-capacity enterprise workloads