TrendAI survey of 3,700 decision makers: AI deployment is outpacing security governance, with agentic AI the main concern
TrendAI™ (Trend Micro) published a global study of 3,700 business and IT decision makers. It finds that organizations are deploying AI faster than they can govern or secure it. 67% of respondents have felt pressured to approve AI despite security concerns, and one in seven called those concerns "extreme" but overridden to keep pace with competitors. Only 38% have comprehensive AI policies, and 57% say AI is advancing faster than they can secure it. The vendor also notes that attackers already use AI to automate reconnaissance, accelerate phishing campaigns and lower the barrier to entry for cybercrime. This is a vendor-sponsored survey, not an incident report, and it contains no vulnerabilities, IOCs or named threat actors.
Detalhes técnicos
- 44% name AI agents accessing sensitive data as their biggest risk.
- 36% cite malicious prompts compromising security.
- 33% cite a growing attack surface, and 33% cite abuse of trusted AI status and autonomous code deployment.
- 31% lack observability or auditability over deployed AI systems.
- Only 48% believe agentic AI will significantly improve cyber defense in the short term.
- Around 40% support AI "kill switch" mechanisms; nearly half are unsure.
- Top-down AI rollouts leave security teams reacting, which drives workarounds and "shadow" AI use.
- 64% report only moderate confidence in their understanding of AI legal frameworks, and 41% cite unclear regulation as a barrier.
Remediação
The article gives no specific technical remediation. It implies the need for stronger AI governance and policy, clear ownership of AI risk, visibility and auditability of AI agents, control mechanisms such as kill switches, and management of shadow AI.
Linha do tempo
-
TrendAI survey of 3,700 decision makers: AI deployment is outpacing security governance, with agentic AI the main concern
ler a análise desta fonte ocultar
TrendAI™ (Trend Micro) published a global study of 3,700 business and IT decision makers. It finds that organizations are deploying AI faster than they can govern or secure it. 67% of respondents have felt pressured to approve AI despite security concerns, and one in seven called those concerns "extreme" but overridden to keep pace with competitors. Only 38% have comprehensive AI policies, and 57% say AI is advancing faster than they can secure it. The vendor also notes that attackers already use AI to automate reconnaissance, accelerate phishing campaigns and lower the barrier to entry for cybercrime. This is a vendor-sponsored survey, not an incident report, and it contains no vulnerabilities, IOCs or named threat actors.
Technical Details
- 44% name AI agents accessing sensitive data as their biggest risk.
- 36% cite malicious prompts compromising security.
- 33% cite a growing attack surface, and 33% cite abuse of trusted AI status and autonomous code deployment.
- 31% lack observability or auditability over deployed AI systems.
- Only 48% believe agentic AI will significantly improve cyber defense in the short term.
- Around 40% support AI "kill switch" mechanisms; nearly half are unsure.
- Top-down AI rollouts leave security teams reacting, which drives workarounds and "shadow" AI use.
- 64% report only moderate confidence in their understanding of AI legal frameworks, and 41% cite unclear regulation as a barrier.
Remediation
The article gives no specific technical remediation. It implies the need for stronger AI governance and policy, clear ownership of AI risk, visibility and auditability of AI agents, control mechanisms such as kill switches, and management of shadow AI.