Shufti launches AI agent verification for authorized commerce

Shufti has launched Know Your Agent, a verification flow for businesses that need to confirm an AI agent’s identity, the human behind it and the permissions behind each request. The product is available now and is aimed at reducing fraud, unauthorized activity and chargeback risk in agentic commerce.

Why it matters: - Agentic commerce is moving AI agents into actions that can open accounts, place orders and move funds on behalf of people and businesses. - Businesses need to verify that an agent is genuine, that a real person approved the action, and that the request stays within assigned limits. - Shufti says the product is designed to help companies accept authorized automation while reducing fraud, unauthorized transactions and chargeback disputes.

What happened: - Shufti launched Know Your Agent (KYA), an AI agent verification solution for autonomous agents acting on behalf of individuals and businesses. - The product is now available for businesses using agentic commerce and other AI agent workflows. - The system verifies AI agent identity, confirms human authorization and checks permissions in a single flow.

The details: - KYA combines agent recognition, trust scoring, spoof-resistant identity verification and permission checks before an agent can access data or trigger a transaction. - Each AI agent gets a unique time-bound credential that can be revoked. - Known agents are recognized by name, while new or unfamiliar agents are assessed through behavioral biometrics. - Each agent receives a trust score to help businesses evaluate risk. - Shufti said it already supports agents including Meta Muse, GrokBot and OpenAI Dots within a week of launch. - Every agent is linked to an individual or business verified through Shufti’s modular KYC or KYB process. - KYA extends those checks to AI agents and does not replace them. - Authorization begins with a principal that sets a mandate for each agent, covering permitted actions, limits, expiry and revocation. - The mandate is enforced as a policy on every request, with ongoing monitoring intended to detect misuse. - Businesses set their own rules inside KYA. - A bank can allow agents to view balances and pay approved billers, while new payees require another check. - A digital assets platform can allow trading within an approved limit while keeping withdrawals blocked. - An iGaming operator can let agents monitor account activity and communicate bet settlement status, while bets must come from the verified player. - A marketplace can support purchases within customer-defined budgets. - A SaaS business can restrict agents to specific resources and tasks. - When a request goes outside a mandate or crosses a high-risk threshold, KYA can require human-in-the-loop approval. - That approval uses a brief selfie liveness check. - Shufti said its single selfie liveness is iBeta Level 3 certified for resistance to presentation attacks. - KYA keeps an audit trail of requests, decisions and revocations tied to each agent and its principal. - The audit trail is meant to keep every action traceable for accountability and review. - The product is also positioned as a revenue tool because businesses can accept authorized agent-driven transactions while blocking rogue activity. - Shufti says this approach lets businesses adopt AI agent automation with human approval over permissions and actions.

Between the lines: - The launch reflects a broader shift from treating AI agents as simple software tools to treating them as actors that need identity, authorization and oversight. - The product leans on layered controls rather than a single login token because stolen credentials and impersonation can make an agent appear legitimate. - Shufti is pitching KYA as both a security control and a commercial enabler for businesses that want to monetize agent-driven workflows without losing control.

What's next: - Shufti is inviting businesses to book sessions to learn how KYA can fit into existing verification and authorization processes. - The company says the system is intended to plug into current workflows for identity verification, human approval and risk management. - As more customer-facing tasks move to AI agents, Shufti is betting demand will grow for tools that balance automation with explicit human control.

The bottom line: - Shufti is trying to become part of the trust layer for agentic commerce by verifying the agent, the owner and the approval chain before money or data moves.