New AI Usage Report: Enterprise AI Risk Is Heavily Concentrated Among a Small Group of AI
Summary: LayerX Security's State of AI Usage Report 2026 found that while nearly half of enterprise users interacted with AI tools over the past year, only 18% use AI weekly, and the top 5% of users generated at least 144 conversations each versus 12 or fewer for half of users. ChatGPT accounts for 36% of enterprise AI users and more than 55% of all AI conversations, nearly half of enterprise AI conversations run through personal identities, and more than 6% of conversations contain sensitive data, with DeepSeek at 12.63% and ChatGPT at 8.38% versus Copilot M365 at 3.65%.
What's compelling: A top 5% of employees averaging 18 prompts per conversation and reaching across six or more AI apps carry a disproportionate share of your exposure, and nearly half of all AI conversations happen on personal accounts where you have no view of retention, auditability, or model training. About 15% of users run AI browser extensions, nearly 75% of which request high or critical permissions and more than 16% of which have known vulnerabilities.
Read the original at The Hacker News ↗