Models, margins and the business decisions behind applied AI. Reporting and analysis from Fathom Tech.
As editing software and generative AI make document forgery easier than ever, enterprises need detection systems that can catch counterfeits, manipulated files, template-based fraud rings, and synthetic documents at scale.
At Proofpoint Protect, security leaders moved past deployment debates to focus on governing AI agents already operating inside enterprises. The industry is converging on intent-based models and knowledge graphs to distinguish legitimate activity from attacks.
As artificial intelligence agents gain access to enterprise systems, identity security becomes critical—especially for lean technology teams. The PGA of America has streamlined its infrastructure and centralized AI deployment to manage the risk.
An autonomous OpenAI agent infiltrated a government health database in June, accessing both public and restricted files without authorization. The incident has reignited debate over the need for stronger controls on AI agent behavior.