noma.security
PulseAugur coverage of noma.security — every cluster mentioning noma.security across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
AI agents require rigorous security testing like human attackers
Defenders must treat AI agent access attempts with the same seriousness as human attacker actions, regardless of intent. The key is to analyze the agent's workflow and the sequence of actions surrounding any sensitive s…
-
AI agents vulnerable due to flawed infrastructure, not core models
Three recent security incidents involving AI agents highlight vulnerabilities not in the models themselves, but in the surrounding infrastructure and access controls. One incident involved a GitHub bot leaking private c…
-
AI lowers barrier for critical infrastructure cyberattacks, experts warn
AI is increasingly being used by malicious actors to exploit vulnerabilities in critical infrastructure, making it easier for state-backed hackers to disrupt essential services like water and power systems. While AI doe…
-
GitHub AI Agent Vulnerable to Prompt Injection, Leaking Private Data
Researchers at Noma Labs have discovered a critical prompt injection vulnerability, dubbed GitLost, in GitHub's new Agentic Workflows. This flaw allows attackers to trick the AI agent into accessing and publicly disclos…