Show Notes
Dan Lisichkin, AI Security Researcher from Pillar Security, discussing their work on "A WIF Of Fresh Access, How a GitHub Issue on Gemini-CLI Led to GCP Project Compromise." Researchers at Pillar Security uncovered a vulnerability in Google’s Gemini CLI workflows that could allow an attacker to turn a malicious GitHub issue into remote code execution and gain elevated access to a Google Cloud Platform (GCP) project.
The exploit chain combined a misconfigured tool allowlist that allowed unrestricted command execution with exposed Workload Identity Federation credentials and overly broad service account permissions, potentially enabling attackers to impersonate privileged accounts. Google promptly addressed the issues by tightening tool restrictions, blocking the agent’s access to credential files, and limiting service account impersonation permissions, highlighting the risks of connecting AI agents to untrusted input and sensitive infrastructure.
The research and executive brief can be found here:
A WIF Of Fresh Access: How a GitHub Issue on Gemini-CLI Led to GCP Project Compromise
Learn more about your ad choices. Visit megaphone.fm/adchoices
More from CyberWire Daily
Today we have Damon Rouse, Threat Intelligence from Lumen's Black Lotus Labs, discussing their work on "The infrastructure quartermaster: inside a China-nexus state enablement model." Black Lotus Labs identified a China-nexus cyber infrastructure provider, dubbed a “quartermaster,” that supplies reconnaissance, proxy routing, and obfuscation services to Chinese cyber espionage actors.
Attackers target critical PHP flaws, and former CISA director Chris Krebs weighs in on AI’s role in the vulnerability surge. Railway confirms a cyberattack. HPE patches critical ArubaOS flaws. Quest Diagnostics investigates a disruptive intrusion. Capcom reports a data breach. China-linked hackers put AI-powered attacks on autopilot. And researchers uncover a Microsoft 365 campaign that hijacks authentication tokens to bypass MFA. Our guest is Anudeep Parhar, COO at Entrust, sharing his thoughts on AI kill switches. eMudra throws in the certificate towel.
Amazon spots PHP exploits before disclosure. GitLab patches a critical flaw. Meta faces a $25 billion penalty. Autonomous AI agents power new cyberattacks. Hackers breach a Japanese cloud provider. CyberCartel unleashes AI against Latin American banks. Tor exit nodes are hijacked. OpenAI coding agents disable security monitoring. Interpol dismantles thousands of cybercrime networks across Africa. Our guest is Danny Manimbo, ISO and AI Practice Leader at Schellman, explaining why companies waiting for AI regulatory clarity may already be behind. Spotify hits the skip button on AI slop.
Critical PHP flaws threaten millions of websites. Weak certificates from an Indian authority spark a global security scramble. Investigators probe a U.S. consulting executive’s alleged ties to ShinyHunters. SafePay ransomware hits at least 600 critical infrastructure organizations. Researchers find troubling security biases in Chinese AI-generated code. A $10 million initiative bolsters cybersecurity for small U.S. utilities. An American pleads guilty to stealing from thousands of servicemembers and veterans. Leo Scott, Managing Director of DataTribe, is talking about their upcoming Cyber Innovation Conference and the 9th annual DataTribe Challenge. Two pallets of GPUs never make it to baggage claim.
A Google Cloud outage disrupts UK banks. DHS bans DeepSeek. Shape-shifting ransomware targets Macs. Bank of America hires an army to patch AI-discovered flaws. Canada looks to bypass the U.S. with a transatlantic cable. A contractor breach exposes California student data. Cybersecurity hiring slumps, but AI skills are booming. Researchers expose the Silent Ransom Group’s playbook. A suspected Qilin ransomware operator takes a vacation straight into custody. Our guest is Roman Arutyunov, Co-Founder of Xage Security, discussing the next era of Zero Trust. A hotel phishing scheme gets a chilly reception.