Cequence Security and KPMG Examine Governance Risks as AI Agents Take on Real-World Tasks
Cequence Security has published a report developed with input from KPMG in Singapore, examining how organisations can govern and secure AI agents as they move beyond generating content to carrying out actions on behalf of businesses and consumers.
Titled Agentic AI Runtime Security, Governance, and Control, the report looks at the security and governance challenges associated with AI agents that can perform tasks such as issuing refunds, updating records and transferring money. Its findings are due to be presented to business, technology and policy leaders at a joint event at KPMG’s Singapore office on November 19.
Data collected by Cequence’s CQ Prime research team during one week in October 2026 identified more than 250,000 AI agent visits across five industries. Approximately one in seven involved agents booking, purchasing, signing in or managing accounts on behalf of customers.
The analysis also found that some agents did not identify themselves as automated systems. At a major telecommunications carrier, traffic from agents that did not disclose their identity reportedly increased from around 100 requests a day to as many as 32,000 within three weeks of a major consumer AI agent launch. These agents signed in using customer credentials and passed multi-factor authentication.
Cequence also identified activity on an online marketplace involving visits claiming to originate from Google-Agent. The number of such visits exceeded the genuine agent visits tracked by the company during the same week. However, all of the visits went directly to the sign-in page and failed, which Cequence said was consistent with attempts to test stolen credentials in bulk. The company clarified that Google was not involved.
The report recommends that organisations assign AI agents clearly defined roles and restrict their authority to specific tasks. It also advocates real-time checks of individual actions by an independent control mechanism, allowing actions to be approved, escalated or blocked based on the agent’s assigned responsibilities.
Cequence argues that security measures should support legitimate task completion rather than focus solely on preventing actions. In a controlled study using a Cequence research prototype, agents operating with logins and fixed permissions attempted actions outside their assigned tasks in 44% of runs. When agents were given defined roles and each action was checked against those roles, the rate fell to 1%. Task completion increased from 76% to 95%, while token usage declined by approximately 35% per run.
“Organizations across Asia-Pacific are moving from experimenting with AI to deploying them to achieve real business outcomes. Traditional controls such as identity verification and fixed access permissions are important, but they may not be sufficient when an agent can carry out a sequence of actions on behalf of a business or customer. Organizations need governance that considers the agent’s role, the context of each action, and the cumulative risk created across an entire task. KPMG’s contribution to this report brings together our experience in governance, risk, regulation and cyber security to help organizations put appropriate oversight around AI agents while enabling legitimate work to continue,” said Gerry Chng, Partner, Head of Cyber, KPMG in Singapore.
“An approval at step one shouldn’t carry into step five,” said Shreyans Mehta, CTO and co-founder of Cequence Security, who co-authored the report. “Runtime governance means checking what an agent does at every step. When the answer is no, it means telling the agent why, so it gets back on track instead of looking for a way around. It has to work the same way for an agent your customer sends as for one you built yourself.”
Cequence said it has established its Asia-Pacific regional headquarters in Singapore and is expanding its regional operations. Its current presence includes a regional team and senior leadership in Australia, with plans to build local teams in the Philippines, Singapore and South Korea. The company lists NTT DOCOMO in Japan and Philippine Airlines among its regional customers.
“AI agents are already acting for customers across the region, and boards want to know they can trust them,” said Ameya Talwalkar, CEO and co-founder of Cequence Security. “KPMG has a deep understanding of how boards and regulators think about risk, and Cequence sees what agents actually do on our customers’ applications and APIs every day. Together, we can show organizations how to give agents real work without losing control. Singapore is setting the pace on governing AI agents, which makes it the natural home for our Asia-Pacific business.”



