White House launched Gold Eagle, linking CISA, open source maintainers, and critical infrastructure operators through AI vulnerability triage under EO
UMKC researchers demonstrated Ghostcommit, a PNG-based prompt injection attack that tricks AI code reviewers into exfiltrating .env secrets as code
Wiz Research’s GhostApproval attack uses symlinks in cloned repositories to trick six AI coding agents into writing attacker SSH keys
Elastic Security Labs found REF6045 deploying SCMBANKER, an AI-written PowerShell toolkit that lets operators control Mexican banking sessions live and
Tel Aviv University and Intuit documented HalluSquatting: AI coding tools hallucinate package names up to 100% of the time, which
Carnegie Mellon researchers found GitHub Copilot refuses harmful prompts 99% of the time in chat but produced harmful code in
Microsoft warned enterprises that its AI-assisted vulnerability discovery tools will produce higher Windows patch volumes and more frequent out-of-band updates.
The UK NCSC published its Cyber Shield blueprint on July 7, outlining autonomous AI agents to discover and remediate vulnerabilities
Sand Security found a one-click session isolation flaw in Writer AI letting attackers access any enterprise tenant’s private models, credentials,
Tokyo police arrested a 15-year-old who used ChatGPT to generate attack code that canceled 46,812 Bandai Channel streaming accounts in
Subscribe to the Daily Security Review Email Briefing to stay informed on the latest threats, trends, and technology, along with insightful columns from industry experts.