Mark RadarMARK RADAR
About
EN
Sign in
Event File AI Generative AI

2026 Enterprise Cybersecurity Survey Flags LLM Data Leaks and Deepfakes as Emerging Threats

5 reports · First detected 2026-04-30 · Last active 2026-05-05

iThome's 2026 Enterprise Cybersecurity Survey examines the system, workforce and AI risks facing businesses, with separate findings for companies with at least NT$10 billion in revenue, general manufacturers, service providers and high-tech manufacturers. The survey found that ransomware and phishing remain core threats, while the rapid adoption of generative AI is expanding the attack surface for data-governance failures and identity impersonation.

The latest 2026 survey found that companies are on high alert for employees leaking sensitive data through LLMs and for deepfake scams impersonating executives or other associated individuals. As agentic AI gradually enters workplace processes, organizations are also focusing on supply-chain vulnerabilities involving third-party models, plugins and data sources, as well as unauthorized access or erroneous actions caused by autonomous systems operating improperly.

All Coverage

5 original reports

The Backstory

The history behind this event

No historical echoes for this signal

Mark Radar|MARK RADAR

If you search news on Google, you can set Mark Radar as a preferred source—our coverage will show up more often in your results. Set as preferred source on Google →

All times are in Taipei time (GMT+8)