Key Takeaways
- Fraud-prevention platform Forter announced Forter Agents, a set of five new purpose-built AI agents, alongside early access to the Forter Model Context Protocol (MCP) that feeds its data into the AI tools merchants already use
- The release extends domains long handled by manual work — payment authorization, chargeback disputes, abuse prevention — toward AI agents that analyze and decide autonomously
- For merchants, the open question becomes how to fold the agentic-era trust and fraud layer into their own decision-making as the next competitive condition
Forter ships its AI agent suite, Forter Agents
On June 24, 2026, Forter, the AI decisioning platform for commerce, announced Forter Agents — five new AI agents — together with early access to the Forter Model Context Protocol (Forter MCP). The agents take over work that teams have run by hand: payment authorization, chargeback defense, and abuse prevention.
Forter launched Forter Agents and early access to the Forter Model Context Protocol for commerce modernization.
www.prnewswire.comForter is a platform that makes an instant trust decision at each commerce moment — from sign-up to checkout to returns. Built on a network of two billion shoppers and data from nearly one million merchants, it stops fraudulent transactions and chargebacks before they happen. This announcement is best understood as adding a layer of autonomous, task-executing agents on top of that decision engine.
The AI agents here differ from a chatbot that acts only when a human prompts it. Each is handed a specific workflow and then collects, analyzes, investigates, and decides on its own. Forter frames the new agents as an extension of its existing conversational suite, Forter Prism, designed around discrete jobs.
Five agents, each owning a workflow
Every newly launched agent is built to address a specific task that merchants otherwise shoulder manually.
At the center sits the Analytics Agent, which surfaces instant insight across risk analysis, transactions, decision rationale, business performance, and customer profiles. It compresses the cycle of gathering data, analyzing it, and writing it up, so teams can react faster. The Dispute Agent then automates the chargeback dispute process. A chargeback is the mechanism by which a cardholder reverses a payment — often citing an unrecognized charge — and for merchants it has long been a source of reversed revenue and manual toil. Automating it aims to recover more of that revenue.
The remaining three carry equally defined roles: the Abuse Agent builds best-practice policies against the abuse of returns, promotions, and loyalty programs; the Payments Agent monitors payment performance and delivers specific recommendations to lift authorization rates; and the Integration Agent connects Forter to existing systems through an AI coding assistant, which Forter says deploys it three times faster.
| Agent | What it does | Problem it solves |
|---|---|---|
| Analytics Agent | Surfaces instant insight across risk analysis, transactions, decision rationale, performance and customer profiles | Slow decisions caused by manual data collection and analysis |
| Dispute Agent | Automates the chargeback dispute process | Manual effort and unrecovered revenue |
| Abuse Agent | Builds policies against returns, promotion and loyalty abuse | Margin and customer-experience erosion from abuse |
| Payments Agent | Monitors payment performance and recommends ways to lift authorizations | Lost revenue from declined authorizations |
| Integration Agent | Connects Forter to existing systems via an AI coding assistant | Slow deployment and delayed time-to-value |
What stands out is that these agents reach beyond conversation. Forter says they turn slow, manual tasks into instant insight and decisions rather than stopping at dialogue. They bring network-backed judgment into domains — authorization-rate improvement, chargeback recovery — that specialist teams have run on experience and intuition.
How MCP support changes the flow of data
The second axis of the announcement is early access to Forter MCP. It pipes Forter's commerce intelligence and identity-network data directly and securely into the AI tools merchant teams use every day, such as ChatGPT and Claude.
The Model Context Protocol is an open standard, introduced by Anthropic in 2024, for connecting AI models to external tools and data sources. It provides a common socket for feeding data to AI, sparing vendors from building bespoke integrations one by one. Forrester projects that 30% of enterprise app vendors will launch their own MCP servers in 2026, and Forter's move fits that current.
With Forter MCP, teams can analyze performance trends, investigate transactions, and generate executive summaries inside the AI tools already embedded in their workflows. Crucially, this happens while preserving the same corporate permissions and security controls — data opens up to external tools without loosening who is allowed to see what.
Opening up the flow of data carries meaning beyond a single feature. Signals accumulated through fraud prevention seep out of a dedicated dashboard and into the AI context that teams face daily. Risk judgment then shifts from the work of a specialist department toward a shared foundation the whole organization can reference.
The trust layer as the defining question of the agentic era
Why would a fraud-prevention company put AI agents and MCP front and center now? Because commerce itself is being rebuilt by AI.
As the speed at which consumers discover, engage, and transact is transformed by AI, the bar for the visibility and intelligence merchants need keeps rising. Fraud, meanwhile, is also weaponizing AI. Forter's network has observed sharp spikes in agent-driven traffic after the debut of ChatGPT Agent, making it a fresh challenge to tell legitimate agents from malicious ones.
This is where network scale matters. Forter CTO Eran Vanounou said that intelligence is now accessible in seconds, in context, with the accuracy a 13-year global network has built. AWS's digital commerce lead echoed that decisions grounded in two billion shoppers give teams the edge to move from signal to action faster than rivals.
In an era where agents step to the front of commerce, the layer that judges whether a counterparty is trustworthy — the trust and fraud layer — grows heavier. Whether to authorize a payment, how to treat identity verification, how to frame a chargeback dispute: Forter is steering these decisions away from human speed and toward autonomous calls grounded in network data.
What it means for merchants
Two points stand out for merchants.
The first is that fraud prevention is ceasing to be a purely defensive function. Forter's agents push into revenue-adjacent territory — recovering chargebacks, lifting authorization rates. Treating risk judgment and revenue optimization on the same data foundation reframes fraud prevention from a defensive cost into an offensive instrument. The question is whether your own fraud program can be discussed in terms of reducing lost revenue.
The second is that data interoperability, exemplified by MCP, is becoming table stakes. Just as Forter opened its data to AI tools, merchants will need to keep their own signals and decision rationale in a form that AI agents — internal and external — can consume. Data locked inside a dedicated tool goes underused in the agentic era. Whether you can stream data into AI workflows while preserving control over who accesses what will determine whether the trust layer becomes a competitive strength.
Conclusion
Forter's announcement of Forter Agents and MCP support illustrates how a fraud-prevention platform is evolving into the trust layer of the agentic era. Five purpose-built agents take on manual domains — payment authorization, chargeback defense, abuse prevention — while MCP lets decision data flow into the AI tools merchants use every day. For merchants, reframing fraud prevention as a tool for cutting lost revenue, and shaping their own signals so AI agents can use them, becomes the precondition for the next round of competition.
