AI Agents Are Now a Mainstream Trigger for Bot-and-Fraud Tooling, Not Just a Security Team Concern

Forrester VP Sandy Carielli’s post-Wave writeup on Bot And Agent Trust Management Software carries a signal buried in the customer references, not the vendor ratings: a majority of surveyed customers are already implementing, or planning to implement, these platforms specifically to monitor inbound AI agent behavior — not just traditional bot traffic. The most common use case remains account fraud, followed by web scraping, LLM scraping, and e-commerce fraud, but Carielli notes interest in these tools has spread well beyond security teams to other business stakeholders. Customers also told her that vendor threat-research strength matters more than perfect detection: they tolerate occasional errors as long as vendors respond fast with model updates and tuning support.

That shift in buying trigger lines up with where the spend is actually headed. Gartner forecasts the market for securing AI will reach almost $4.8 billion in 2027, a 68.7% increase over 2026, with AI usage control — governing how employees and agents actually use AI tools day to day — the fastest-growing segment at 73% growth. Gartner’s own warning sharpens the stakes: by 2029, more than half of successful cyberattacks on AI agents will exploit access-control weaknesses and prompt injection specifically, naming agentic AI deployment a distinct new attack surface rather than an extension of traditional application security.

Read together, the two reports describe the same market forming from opposite directions: Forrester’s customer references show buyers already shifting bot-and-fraud budget toward monitoring agent traffic, while Gartner’s forecast shows the category growing fastest exactly where that monitoring has to happen — usage control, not model security. For B2B sellers, that’s a live buying trigger, not a future one: security and GTM stakeholders are already in the room together on this purchase.