Agentic AI & LLMs: Top Cybersecurity Concerns for Pros

A new report reveals that agentic AI, large language models, and cloud infrastructure breaches are top concerns for cybersecurity professionals. The 2026 Cybersecurity Assessment Report surveyed over 1,200 professionals across six countries. Internal AI systems and large language models are the primary environmental concerns for 45% of respondents. Cloud infrastructure and applications follow closely at 44%. In the past year, 41.8% of respondents experienced cloud infrastructure or application breaches. Business email compromise affected 35.9%, and ransomware impacted 25.6%. A significant 59.2% of respondents confirmed AI-driven social engineering attacks. The top AI-driven threat is attackers using AI for self-mutating malware, cited by 55.9%. This information is crucial for understanding the evolving landscape of cyber threats.