Bitdefender: Managers Overestimate AI Security Visibility

A new report reveals a significant disconnect in how organizations view their AI security. Managers believe they have a clear understanding of how staff use artificial intelligence, but the analysts who manage the tools often disagree. This confidence gap is evident across multiple findings. For example, 57.8% of managers report full visibility into AI use, compared to only 45.9% of their security practitioners. Nearly half of organizations admit they only have partial visibility into "shadow AI" and personal large language model accounts on their networks. The report also highlights a concerning trend: 55.2% of respondents breached in the past year were told to keep the incident quiet. While managers are concerned about advanced AI threats, attackers are primarily using AI to enhance current campaigns. This overconfidence in AI security could leave organizations vulnerable to unmonitored internal AI use.