AI Briefing#cybersecurity
The latest AI and developer news about #cybersecurity, with the original source and a short summary.
The latest AI and developer news about #cybersecurity, with the original source and a short summary.
OpenAI has released the safety report for GPT-5.2-Codex, an agentic coding model optimized for complex software engineering.
As AI's cybersecurity capabilities improve, OpenAI is strengthening a multi-layered security strategy to support defenders and prevent misuse.
Amazon's ATA uses AI agents to simulate and validate threats, automatically improving detection.
Anthropic detected and disrupted the first large-scale cyber espionage campaign that exploited AI agent technology.
OpenAI has unveiled Aardvark, a security agent built on GPT-5 that detects and even patches code vulnerabilities.
Doppel used GPT-5 and RFT to cut cyberattack response time from hours to minutes.
OpenAI released a report detailing cases in which it detected and blocked malicious uses of its AI models over the past quarter.
KakaoBank practically supported the participation of Cold Fusion, a Korean white-hat coalition team, in DEF CON 33 CTF, fulfilling a DevRel role.
OpenAI announced an outbound coordinated vulnerability disclosure policy for responsibly reporting vulnerabilities it discovers in third-party software.
Replit has strengthened data exfiltration blocking and security scanner features to protect users from the 'Shai-Hulud' worm attack.
Anthropic has established a National Security and Public Sector Advisory Council to help the U.S. government and allied nations secure a technological edge.
This records the challenges experienced while participating in the DEFCON 33 CTF finals.
IGLOO implemented the on-premise LLM security assistant AiR using Rebellions NPUs.
Malicious fine-tuning of gpt-oss was used to assess the biological and cybersecurity risks of open-weight LLMs.
Outtake is an AI agent powered by OpenAI's GPT-4.1 and o3 that has increased the speed of digital security threat detection and response by more than 100x.
Anthropic donates $2 million to Carnegie Mellon University to support energy and cybersecurity infrastructure, aiming to secure America's AI competitiveness.
OpenAI announced results of using AI to detect and block AI misuse cases such as social engineering and cyber espionage.
Replit has introduced enhanced security features, including authentication, security scanning, and file protection, to address the security threats of AI-powered 'vibe coding.'
Replit provides core security principles like version control and encrypted storage to prevent vulnerabilities in AI-generated code.
OpenAI is significantly expanding its cybersecurity research support and bug bounty rewards to prepare for the AGI era.
OpenAI has released a report detailing the latest cases and response status for detecting and preventing malicious use of AI models.
As AI advances rapidly, targeted regulation that does not stifle innovation is needed to address cyber and CBRN risks.
OpenAI has disrupted more than 20 deceptive operations since earlier this year and has released its latest threat intelligence report.
OpenAI operates an independent board-level Safety and Security Committee to oversee the safety of model development and deployment.