Firms deploy AI faster than they can manage associated risks: Report

Bengaluru: Many firms are deploying AI under pressure, and the risk of pressure-driven AI rollout is exacerbated by governance inconsistencies and unclear responsibility for AI risk that are becoming widespread.

A new global study of 3,700 leaders, including 200 business and IT decision-makers in India, by AI security company TrendAI revealed that nearly 80 per cent of respondents in India have felt pressured to approve AI despite security concerns, with only 42 per cent of IT managers and 49 per cent of business managers feeling moderately prepared for the pace of IT adoption.

The report said that many organisations are still determining their compliance obligations related to AI, and this is challenging in a fragmented global regulatory environment that continues to evolve.

It said that in India, this uncertainty is particularly pronounced because preparedness levels remain low even among senior decision makers. Only 49 per cent of business decision makers and 42 per cent of IT decision makers are moderately prepared for the pace of AI adoption.

The research also revealed fragmented ownership of AI risk in India, as among IT decision makers, there is no clear consensus on who is accountable for AI risk management.

It said with accountability unclear, security gaps are more likely to go unaddressed and pressure to move fast only compounds the problem. About 80 per cent of respondents feel pressured to approve AI implementations that posed potential security risks.

Resolving vulnerabilities after deployment is far more costly than addressing them earlier in development, and it also introduces new risks to firms.

The report also mentioned that AI-powered cybersecurity presents an opportunity for Indian firms to strengthen defences against increasingly sophisticated threats.

Though interest in AI-driven capabilities is high, skepticism around agentic AI systems remains. IT decision makers cite malicious prompts as their top concern, and business decision makers worry most about the expanded attack surface for cybercriminals.

Sharda Tickoo, Country Manager, India & SAARC, TrendAI said, “Organisations are not lacking awareness of risk. They are lacking the structures to act on it. Fragmented ownership, pressure- driven deployment, and limited expertise are creating conditions where AI becomes embedded in critical systems before the controls exist to manage it safely. As organisations move toward more autonomous AI, that gap becomes far harder to close."

The report has recommended that firms should define clear governance structures before scaling AI deployment, and this includes assigning ownership for AI risk, setting acceptable use policies and ensuring alignment with evolving regulatory requirements.

The report also highlights the need to understand where AI systems are operating across the organisation.