News
ai An updated Model Spec for more capable and beneficial AI - OpenAI - ChatGPT & AI Research cybersecurity Tracking AI adoption in North Korean threat actor tradecraft - Microsoft Security Blog cybersecurity Linux Kernel TCP Flaw With 9.8 CVSS Score Affects Ubuntu, Red Hat and Other Distributions - The Hacker News cybersecurity Threat Modeling With 4 Simple Questions on a White Board - Dark Reading cybersecurity Fake ChatGPT and Claude Apps Abuse Signed Executables to Deploy Crypto Miners - The Hacker News threat-intel Coruna iOS Exploit Kit Hidden Behind Malicious Ads Targets Crypto Seed Phrases, ADEX Reports - GBHackers cybersecurity Google Says Gemini 4 Carbon Leads in Finding, Exploiting and Fixing Vulnerabilities - SecurityWeek cybersecurity How Secure Are US Elections? Voting Machine Risks and the Growing Threat of AI - SecurityWeek cybersecurity Cisco Open Sources Quarry to Catch AI Agent Attacks Hidden in Harmless Steps - SecurityWeek cybersecurity CISA warns of Fragnesia Linux flaw chained in PHP2Shell attacks - BleepingComputer threat-intel AppViewX Expands Quantum Trust Hub with Cryptographic Discovery, Posture Management, and Remediation - GBHackers cybersecurity Adobe, Veeam, Juniper Networks Patch Critical Vulnerabilities - SecurityWeek cybersecurity Google's Gemini 4 Carbon Leads Benchmarks for Finding, Exploiting, and Patching Vulnerabilities - The Hacker News threat-intel Cisco Talos Uncovers KeyDrift Malware Targeting SD-WAN Controllers to Steal Network Encryption Keys - GBHackers cybersecurity BAE Says No Data Extracted in Incident Hitting Bankruptcy Courts - SecurityWeek
1 / 15
All news ›
Cybersecurity

Google's Gemini 4 Carbon Leads Benchmarks for Finding, Exploiting, and Patching Vulnerabilities

Google on Wednesday took the wraps off Gemini 4 Carbon, its latest frontier model with state-of-the-art cybersecurity and long-horizon capabilities. "It shows significant gains across coding and agentic tasks as well as across key capabilities critical for cybersecurity defense, landing a firm lead in SEC-bench Pro (88.2%) for finding vulnerabilities, ExploitGym (54.8%) for exploitation, and

Continue on The Hacker News ↗ ‹ All news Get the newsletter

The summary and image come from The Hacker News; the image is shown from their website.