Riskified, a global leader in ecommerce fraud prevention and risk intelligence, has formed a strategic partnership with HUMAN Security, a renowned cybersecurity company. Together, they aim to create a unified security framework that empowers merchants to grow confidently in the era of emerging AI-powered shopping agents. This collaboration leverages both companies’ industry-leading AI platforms and vast network insights to secure the next generation of digital commerce.

Although fully autonomous shopping agents have yet to become mainstream, consumer behavior is evolving rapidly. Increasingly, shoppers are turning to large language models (LLMs) like ChatGPT, Claude, Gemini, Grok, Llama, and Perplexity to research products, compare prices, and hunt for deals. Recent announcements from LLM leaders about integrated browser experiences signal even greater adoption ahead. While this shift offers merchants exciting opportunities to capture new customers and boost conversions, it also introduces significant risks. Traditional rules-based fraud detection often struggles when AI agents handle transactions, as these interactions bypass critical behavioral signals—leading to higher false declines or undetected fraud.

AI Authority Trend: Synack Unveils Agentic AI Architecture with Human-in-the-Loop for PTaaS

Without a strong trust framework, merchants risk losing revenue, facing inventory abuse, and suffering damage to their reputation. Riskified’s early data shows that LLM-referred traffic can be substantially riskier than traditional sources. For instance, traffic from a major ticketing merchant proved 2.3 times riskier than Google search referrals, while an electronics merchant saw 1.8 times higher risk. Moreover, Riskified has detected early signs of automated reseller arbitrage, where AI agents rapidly deplete inventory and resell products through fraudulent storefronts often promoted by other AI agents. If left unchecked, such practices could disrupt pricing strategies, undermine customer trust, and inflict severe financial losses.

To tackle these challenges directly, Riskified is introducing new tools in partnership with HUMAN Security:

  • AI Agent Approve is an MCP Server package available on the AWS Marketplace that allows merchants and LLM platforms to securely connect with Riskified’s APIs, helping ensure safer adoption of AI-powered shopping.
  • AI Agent Intelligence: A specialized dashboard within the Riskified Control Center that lets merchants monitor ecommerce orders placed by AI agents.
  • AI Agent Policy Builder: Tools to detect and prevent agentic policy abuse, such as reseller arbitrage, programmatic returns abuse, and promo exploitation.

AI Authority Trend: AKOOL Unveils Holographic Avatar to Transform AI Human Engagement

By combining HUMAN’s newly launched Sightline with AgenticTrust and Riskified’s expertise in fraud prevention, chargeback protection, and policy abuse mitigation, the partnership enables merchants to apply consistent trust policies across both human and AI-driven transactions. This joint approach aims to approve more legitimate AI-powered orders, block sophisticated fraud, and protect customer relationships.

Assaf Feldman, CTO and Co-Founder of Riskified, highlighted that establishing identity and trust in AI-driven transactions remains a complex challenge. “By working with HUMAN, we help merchants turn potential threats into profitable new channels,” he said. John Searby, Chief Strategy Officer at HUMAN Security, added, “Together, we set the standard for safe, profitable agentic commerce by combining HUMAN’s trust layer with Riskified’s fraud prevention expertise.”

AI Authority Trend: Human Energy Grid: DebitMyData Secures Funding for Global Rollout

To share your insights, please write to us at sudipto@intentamplify.com