Almost all AI tools are now running with no oversight from IT — putting companies in the firing line (www.techradar.com)

A recent report by Reco reveals that a startling 80% of AI tools within organizations are operating without any oversight from IT departments, raising significant concerns about cybersecurity and data security. The analysis included 500 Model Context Protocol servers and highlighted 637 vulnerabilities across various AI agents and large language models (LLMs). Notably, smaller companies are at higher risk, utilizing an average of 414 AI tools per 1,000 employees without proper approval, often through unmonitored browser extensions and workflows. This lack of IT governance presents alarming operational risks, as many AI tools can access both local data and the internet, creating opportunities for data exfiltration. The report’s findings stress the urgency for organizations to resource their IT teams adequately in order to manage unapproved AI usage. With AI agents now deeply integrated into daily business processes, the absence of oversight not only jeopardizes sensitive data but also exposes companies to new vulnerabilities, underscoring the need for a reevaluation of AI tool deployment policies and security measures.