Volume 343, Issue 2 · 6 articles from 4 issues
TrendAI previews an agentic service that finds, validates and verifies fixes for vulnerabilities, including zero-days, in source code.
TrendAI announced a private preview of TrendAI Vision One Autonomous Vulnerability Discovery, powered by its agentic exploit-remediation engine, code-named AESIR.
The multi-model system, built primarily on Anthropic models, identifies, validates and verifies fixes for vulnerabilities including zero-days in customer source code.
It combines autonomous discovery and threat hunting with research from the Zero Day Initiative bug bounty program, outperformed entries on the CyberGym leaderboard and runs on Amazon Bedrock.
TrendAI research shows how national AI ecosystems shape AI adoption by state-aligned APTs in China, Russia, North Korea and Iran.
TrendAI research summarizes its report 'AI Statecraft and Cyberthreats in 2026' on how national AI ecosystems shape state-aligned APT activity, finding China best positioned to integrate AI into cyber operations thanks to redundant domestic models.
Russia-aligned actors have moved to operational use, including LLM-generated malware commands and AI-generated decoy code.
North Korea relies on foreign AI services and its IT worker scheme, which has infiltrated more than 100 American companies, and cross-actor cooperation may spread AI capabilities between states.
TrendAI Vision One now ingests Claude Code and Cowork session records so teams can investigate AI agent activity.
TrendAI expanded its TrendAI Vision One support for Claude's Compliance API to cover Claude Code and Cowork session transcripts from Claude Enterprise, building on an integration announced in June 2026.
The records include prompts, responses and tool calls, each tied to a verified user.
Security teams can investigate AI agent activity in context and correlate it with endpoint, identity, cloud and network telemetry to see what agents were asked to do and whether it followed policy.
Trend Micro's CEO argues AI agents can be secured using established cybersecurity practices, not rejected outright.
Trend Micro CEO Eva Chen argues that AI agents can be secured through established cybersecurity practices rather than rejected outright.
Drawing a parallel to the development of electricity, she contends engineers can implement safeguards such as identity authorization, isolation, monitoring, and policy controls to govern how AI systems act.
The piece argues this allows organizations to harness AI's benefits while maintaining safety and accountability.
TrendAI data shows Japan logged more disclosed cyber incidents in nine months of 2026 than in all of 2025.
TrendAI issued a threat advisory for organizations in Japan after a senior Liberal Democratic Party official declared that cyberspace is in a state of emergency.
TrendAI Forward-Looking Threat Research data shows Japan recorded more publicly disclosed cybersecurity incidents in the first nine months of 2026 than in all of 2025.
September alone saw 86 incidents, up about 18% from August and 37% from July, with breaches hitting retailers, financial services firms and other businesses.
Pwn2Own Ireland participants found 98 unique zero-days, earning $1,262,000 from TrendAI's Zero Day Initiative.
TrendAI's Zero Day Initiative awarded $1,262,000 at Pwn2Own Ireland, where contestants discovered and disclosed 98 unique zero-day vulnerabilities over three days.
Targets spanned mobile phones, smart home devices, wellness products, printers, messaging, AI infrastructure and AI coding agents.
The vulnerabilities go to vendors for patching, and TrendAI says visibility into them also protects its customers through virtual patching.