Zenity Labs disclosed three zero-click vulnerabilities in Salesforce Agentforce, collectively dubbed SalesBleed, in which a prompt injection hidden in a public Web-to-Lead form caused the General CRM …
Zenity Labs disclosed three Salesforce Agentforce vulnerabilities, collectively named SalesBleed, that let attackers plant indirect prompt injections through the Web-to-Lead form and make AI agents ex…
At Black Hat USA 2026, Zenity Labs demonstrated that every major AI browser—Claude in Chrome, Gemini in Chrome, Perplexity Comet, ChatGPT Atlas, and Copilot in Edge—can be fully hijacked without a sin…
A new class of security threat called 'agent swarm attacks' is emerging, where multiple AI agents combine individually reasonable actions into damaging outcomes without a single malicious actor, as hi…
Zenity Labs discovered AgentForger, a phishing-based attack that silently creates and launches fully autonomous AI agents within OpenAI workspaces with access to Outlook, Slack, SharePoint, and Google…
Zenity Labs researchers discovered AgentForger, a phishing-based attack that silently creates and launches a fully autonomous AI agent within OpenAI workspaces, giving attackers persistent access to a…
Security firm Zenity Labs disclosed a vulnerability in OpenAI's ChatGPT Workspace Agents that allowed a single manipulated URL to silently create an autonomous AI agent inside an enterprise environmen…
Zenity Labs uncovered 'AgentForger,' a vulnerability in OpenAI's Agent Builder that allowed a single manipulated ChatGPT link to create an autonomous agent on an employee's behalf, inheriting the vict…
Zenity Labs publicly disclosed AgentForger on July 23 after OpenAI fixed a ChatGPT Workspace Agents vulnerability on June 8, four days after Zenity reported it via Bugcrowd. The flaw allowed a weaponi…
Security firm Zenity Labs discovered a flaw in OpenAI's ChatGPT workspace agents, dubbed 'AgentForger,' that allowed one click on a ChatGPT link to silently create a malicious AI agent inside a victim…
Zenity Labs revealed that attackers are weaponizing AI infrastructure, targeting exposed model gateways and unmanaged LLM endpoints. The firm's sensors detected thousands of real-world attacks, includ…