The pioneering verification framework that connects AI brokers to actual, verified human id, redefining digital belief and AI fraud prevention.
Sumsub, a world verification and fraud prevention chief, introduced the launch of its AI Agent Verification answer, a first-of-its-kind strategy to controlling AI-driven automation by binding it to actual, verified human id inside Sumsub’s Know Your Agent (KYA) framework.
As AI brokers and browser-based automation develop in recognition, companies battle to inform respectable exercise from fraud. Most platforms deal with automation as inherently suspicious and block it by default. In mild of this, Sumsub’s AI Agent Verification allows companies to separate lawful, human-driven automation from malicious agent assaults by linking all exercise to a verified human id. This creates a transparent line of accountability, permitting respectable automation to function whereas stopping illicit exercise.
AI Agent Verification treats automation as a manageable threat fairly than an computerized purple flag. The system first detects when exercise is automated, evaluates its threat stage, and solely applies further checks when obligatory. In higher-risk eventualities, AI Agent Verification can require a focused liveness check to substantiate that an actual human is current and approved. By doing so, it prevents deepfakes from getting used rather than actual customers and ensures that each motion is immediately linked to the particular person accountable. The chance-based strategy prevents pointless friction for respectable customers, whereas preserving sturdy safeguards towards abuse.
Additionally Learn: AiThority Interview with Zohaib Ahmed, co-founder and CEO at Resemble AI
“AI brokers are quickly changing into the spine of digital operations, but most of at the moment’s techniques nonetheless deal with them as opaque, unaccountable black containers,” mentioned Vyacheslav Zholudev, Co-founder and CTO at Sumsub. “With AI Agent Verification, Sumsub is the primary to bind AI brokers to verified human identities at scale. Reasonably than making an attempt to blindly belief AI brokers themselves, our answer focuses on verifying the people behind them.”
AI Agent Verification allows risk-based management by constructing on the core capabilities of Sumsub’s full-cycle verification platform:
- Gadget Intelligence and Bot Detection: Detects when exercise is automated and assesses its threat stage in actual time.
- Mule Community Prevention: Analyzes machine habits and network-level indicators past fundamental IP monitoring, uncovering suspicious patterns throughout gadgets, accounts, and periods to reveal coordinated mule exercise earlier than it may possibly scale.
- Liveness Verification: Confirms that an actual human is current and authorizing the agent’s actions at important moments, similar to onboarding, account management adjustments, or high-value payouts.
- Threat Scoring and Monitoring: Repeatedly evaluates behavioral and contextual indicators throughout the client lifecycle, so automation might be allowed, restricted, or challenged primarily based on threat.
In accordance with Sumsub’s Id Fraud Report 2025–2026, AI fraud brokers are rising as a brand new type of evasion as id fraud reaches unprecedented ranges of sophistication, with a 180% year-on-year improve in multi-step, coordinated assaults globally in 2025. As AI brokers and browser-based brokers are more and more deployed for mass payouts and automatic transactions in sectors similar to fintech, funds, e-commerce, and ticketing, automation is changing into a regular lever for effectivity and is now successfully unavoidable.
“As we speak, automation itself isn’t the issue – anonymity is. When AI brokers can autonomously transfer cash, create accounts, or transact at scale with no actual particular person behind them, fraud can nearly turn out to be inconceivable to mitigate,” mentioned Artem Popov, Head of Fraud Prevention at Sumsub. “AI Agent Verification adjustments that dynamic by requiring human accountability on the moments the place automation turns into harmful. Companies have to be assured that if an agent takes motion, there’s at all times an actual, verified particular person chargeable for it.”
With AI Agent Verification, Sumsub helps organizations struggle AI-driven fraud by verifying AI brokers and making certain solely trusted automation can act on their behalf.
Additionally Learn: The Demise of the Questionnaire: Automating RFP Responses with GenAI
[To share your insights with us, please write to psen@itechseries.com]
