Claude Models Attacked Companies in Testing
Anthropic last week disclosed that three of its Claude models gained unauthorized access to real organizations during cybersecurity…
How Can CISOs Prepare for AI Outpacing Defenses in “Months”?
When five of the world’s most powerful intelligence agencies agree on something, the security community listens. In June,…
Microsoft Debuts First Cybersecurity AI Model MAI-Cyber-1-Flash
Microsoft has launched MAI-Cyber-1-Flash, its first AI model designed specifically for cybersecurity, alongside an expanded version of its…
Google Previews Gemini 3.5 Flash Cyber to Lower AI Security Cost
Alphabet has released three new Gemini models, including a cybersecurity-focused system designed to identify and patch software vulnerabilities.…
OpenAI Launches Presence Amid AI Agent Safety Concerns
OpenAI has today launched OpenAI Presence, an enterprise product designed to help organizations deploy reliable AI agents. As agents take on…
Microsoft’s New MCP Server Is Solving Retail’s Security Blind Spot
Microsoft has introduced a managed MCP server in Dynamics 365 Commerce to ensure AI agents interact securely with…
Anthropic Briefs Central Banks on Mythos Vulnerabilities
Anthropic has agreed to brief finance ministries and central banks on vulnerabilities identified by its new Mythos model…
Who Is Accountable When Public Sector AI Goes Wrong?
When a local authority deploys AI to manage transport networks or answer resident queries, responsibility for that system’s…
Securing the Enterprise in the Age of AI Agents
Agentic AI is moving from a promising productivity tool to a security problem that enterprise leaders can no…
ServiceNow AI Governance Push: Knowledge 2026
ServiceNow has dropped three announcements that, taken together, amount to the clearest signal yet of what AI Control…
Cisco Debuts Model Provenance Kit to Audit Third-Party AI Models
Cisco has unveiled a new open source initiative aimed at tackling enterprise AI model procurement. The company’s newly…