OpenAI has given Japan’s three largest banks access to its GPT-5.5 Cyber model to strengthen cybersecurity defenses. The model can perform autonomous penetration testing and code validation, according to reports.
Integrated Cyber Solutions has begun the patent process for MASQ, a governance and security framework designed to control permissions, data access, and reasoning protection in autonomous AI agents.
Humanix has introduced new detection and response capabilities that identify live attempts to coerce support staff into bypassing organizational safeguards during sensitive IT workflows.
SAFE has launched SAFE AI Security Posture Management, a platform designed to give enterprises continuous visibility and control over AI-related cyber risks across their ecosystems.
TrendAI, the enterprise cybersecurity arm of Trend Micro, announced new strategic initiatives focused on organizational evolution, platform enhancement, and partner ecosystem development. The company will present its AI governance and asset visibility solutions at COMPUTEX 2026.
E Tech Group has reported a 400 percent increase in demand for its IT and OT assessments as manufacturers prepare their infrastructure for modernization and AI initiatives. The company provides vendor-neutral evaluations focusing on cybersecurity, system health, and network architecture to help reduce downtime and guide remediation efforts.
MicroAlgo Inc. has announced a quantum encryption technology that integrates lattice cryptography with its LSQb algorithm to enhance resistance against quantum attacks and improve secure information transmission.
Shadow Sciences Group has published a six-part investigative series titled 'When Trust Becomes the Attack Surface,' examining the erosion of identity verification reliability due to AI and proposing a new model for continuous verification.
IBM and Red Hat have announced Project Lightwell, a $5 billion initiative to create a trusted enterprise clearinghouse for open source software security, supported by more than 20,000 engineers and advanced AI capabilities.
Cato Networks has appointed Tristan Elder as Vice President of EMEA Channel to expand its partner program and strengthen its regional channel strategy. Elder previously led the EMEA channel business at ExtraHop Networks and was part of CrowdStrike's early leadership team.
Anthropic said that its Project Glasswing partners, using the Claude Mythos Preview model, have identified more than 10,000 high or critical vulnerabilities in critical software systems within the first month of operation.
Integrated Cyber Solutions has released an updated white paper on its VEIL privacy-preserving machine learning system, reporting data compression above 95 percent while maintaining model performance across enterprise use cases.
Copia Automation has launched Copia Actions, a configurable workflow engine designed to automate code management tasks across PLC, HMI, and robotic controllers in industrial environments.
Crew Scaler has released a detailed 120-page analysis on the security of multi-agent AI systems, evaluating 16 frameworks and identifying major gaps in current safety practices.
CIQ has announced a partnership with Binarly to integrate binary-level and firmware vulnerability detection into Enterprise Linux environments, combining CIQ's hardened OS with Binarly's Transparency Platform for deeper supply chain assurance.
Venture capital firm JVP announced four portfolio company exits in the first quarter of 2026, including sales involving DealHub, Pyramid Analytics, Covera Health, and 1touch.io, reflecting strong returns across its cybersecurity and AI investments.
NeuralTrust has been recognized as a Star in the MarketsandMarkets Agentic AI Security Market report, placing it among the leading vendors in AI security for agentic systems.
Check Point Software Technologies' 2026 Cloud Security Report reveals that while 77% of organizations have updated their cloud security strategies for AI, only 26% can enforce them, exposing a significant gap between security intent and capability.
TELUS Digital has released a benchmark study analyzing the safety of 34 AI models through more than 620,000 adversarial tests, showing that smaller models are more vulnerable and reasoning models are harder to exploit.
Silicom Ltd. has secured a design win worth $5 million per year with a global cybersecurity company for a new range of white label switches, with production orders expected in 2026.
Weekly newsletter about AI in Cybersecurity.
