AI agents are claiming your free trials. Let the right ones in.
AI agents now sign up and claim free trials for real people. How to stop scripted trial farming without turning away the AI agents your customers use, with one setting per benefit.
AI agents now sign up for products, start free trials and claim credits on behalf of real people. A customer asks their assistant to "set up a trial of the three best tools for this" and an agent does the clicking. At the same time, the people farming your free trials use automation too, to create accounts by the hundred.
Both arrive as automation. Turn all of it away and you lose real customers along with the farms. Let all of it through and your trials, credits and free plans get claimed again and again.
Automation doesn't sleep. A script can create accounts and claim credits all night, hundreds every hour, while your team is offline. If each claim carries $2 of free credits, that's thousands of dollars of model spend before anyone wakes up, and a free tier that's slow or down for the real users who arrive in the morning.
From our early beta
A team whose cloud bill normally ran $200 to $300 a month woke up to a $12,000 bill. Automated registrations had triggered thousands of expensive serverless functions, database operations and background jobs overnight.
A bot used one team's phone verification to send 100,000 one-time passcodes. With no bot protection in place, the team paid for every one of those text messages.
This guide shows how to block bots and allow AI agents on your free trials, credits and rewards, with one setting per benefit.
Two kinds of automation
| Helpful AI agent | Scripted trial farming | |
|---|---|---|
| Acts for | One real person | One person, many accounts |
| Claims | One trial per customer | As many as the script can create |
| What it's worth to you | A real customer reached through a new channel | Cost with no customer behind it |
| What you want | Let it through, under your rules | Stop it |
The difference isn't automation itself. It's whether the automation is abusive, and whether the person behind it has already received the benefit.
Why blanket bot blocking backfires
The classic answer to automation is to block it: challenge every signup, turn away anything that doesn't look like a person clicking. That made sense when nearly all automation was abusive. It doesn't anymore.
- It turns away the customers who use agents. Browser assistants and AI agents are becoming a normal way to try software. A blanket block tells those customers your product isn't for them.
- Challenges get solved anyway. Determined farms pay for challenge solving or use real people to pass them. The friction lands on everyone except the people you meant to stop.
- It misses humans claiming twice. Most trial abuse isn't scripted at all. A real person with a second email passes every "are you human?" test. We cover that gap in abuse prevention vs fraud detection vs identity verification.
One setting per benefit: Automation and AI agents
The right treatment of automation depends on what it's claiming. A trial you can end is a different risk from a pack of credits that's spent the moment you grant it. So every AbuseZero benefit has one setting, Automation and AI agents, with three options:
| Option | What it means |
|---|---|
| Allow safe automation and AI agents (the default) | Automation and AI agents can claim this benefit. AbuseZero Automation Defense denies automation that is abusive. |
| Allow verified AI agents only | AI agents verified by AbuseZero can claim this benefit. AbuseZero denies all other automation. |
| Deny all automation | AbuseZero denies claims from automation, including verified AI agents. |
You choose the option in Benefits → Edit benefit, beside Protection. There's no agent list to maintain and nothing to register. The details are in bots and AI agents.
A sensible starting point:
| Benefit | Automation and AI agents | Why |
|---|---|---|
| Free trial you can end | Allow safe automation and AI agents | Agent signups are welcome, and a trial can be ended |
| Generous AI or API credits | Allow verified AI agents only | Credits are spent the moment you grant them |
| Free-forever plan | Allow safe automation and AI agents | Real users reach you through assistants |
| Promotion or referral reward | Deny all automation | Scripted sign-ups are how referral loops scale |
Referral and promotion rewards attract the most scripted abuse. See how to stop promo code and referral abuse.
Where the setting sits in your policy
The setting is one step in how your policy reads. Allow & block lists run first, then the benefit's automation and AI agents setting, then your allow rules, then your deny rules, then AbuseZero protection and the risk score, then the default result.
An allow list entry still lets one person through, whatever the setting says.
What AbuseZero Automation Defense handles for you
AbuseZero Automation Defense runs inside every evaluation, in the same call that checks the claim. Your server makes one request and gets one answer: allow or deny.
- Every claim, inside the time limit. Automation Defense works on every evaluation, together with identity, device, network and benefit history, and the result comes back within the evaluation's time limit.
- AI agents verified by AbuseZero. AbuseZero verifies AI agents itself. You don't register agents, keep a list or update anything when new agents appear.
- Every claim still goes through the rest of your protection. Allow & block lists, your protection level, your custom policies and eligibility apply to every claim, including claims from verified AI agents. An agent claiming for someone who already had the trial is evaluated like any repeat claim.
- A reason you can show your team. When Automation Defense decides a claim, the reason code is
automation: "AbuseZero Automation Defense denied this claim."
Behind every evaluation sits the AbuseZero AI Risk Engine, a proprietary algorithm built on 140,000+ risk signal combinations with AI that checks the claim again. Read more in bots and AI agents.
Welcome agents, stop farms
Choose how each benefit treats automation and AI agents, then watch the results in observe mode before anything changes for users. Start with Builder or see pricing.
Add it to your claim route
Two calls cover it, the same two as any AbuseZero integration.
In the browser, call observe() when the person submits the claim, and send its deviceToken with the claim.
import { observe } from "abusezero/browser";
export async function submitTrial(claimId: string, email: string) {
const { deviceToken } = await observe({ key: "pk_…", claimId });
await fetch("/api/trial", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ claimId, email, deviceToken }),
});
}On your server, evaluate the claim before you start the trial, and send the incoming request with the evaluation. The AbuseZero Node.js SDK takes the Request your route received as it is.
import { abusezero } from "@/lib/abusezero";
export async function POST(request: Request) {
const body = await request.json() as { claimId: string; email: string; deviceToken: string };
const evaluation = await abusezero.evaluate({
benefit: "pro_trial",
claim_id: body.claimId,
subject: { email: body.email },
identifiers: { device_token: body.deviceToken },
}, `evaluate:${body.claimId}`, { request });
if ("status" in evaluation || evaluation.result === "deny") {
return Response.json({ started: false });
}
// Start the trial in your application here.
return Response.json({ started: true });
}Your route keeps one code path for people and agents alike. Then choose the option on each benefit in the dashboard, or send config.automation as allow_safe_automation, allow_verified_agents or deny_all_automation from the AbuseZero REST API, Node.js SDK, CLI or MCP server. Every way to set it is in programmatic access. For where this route sits in your flow, see add a trial and credit abuse check to your signup flow.
Roll it out without surprises
Start in observe mode. Every claim is allowed while AbuseZero records what it would have decided, including for claims from automation.
Keep agent-friendly benefits on Allow safe automation and AI agents, the default, where agent signups make sense for your product.
Tighten where it's expensive. Set generous credit packs to Allow verified AI agents only, and promotion and referral rewards to Deny all automation.
Switch to protect when the results look right, and keep an eye on denials as agent traffic grows.
Deny all automation also denies verified AI agents, so keep it for benefits no agent should claim, and look at the effect in observe mode first. More on keeping real users safe in multi-accounting detection without blocking real users.
Frequently asked questions
Should I block AI agents from claiming my free trial?
Not all of them. AI agents increasingly act for real customers, so a blanket block turns those customers away. Keep agent-friendly benefits on Allow safe automation and AI agents, choose Allow verified AI agents only where each claim costs more, and save Deny all automation for benefits no automation should claim.
How do I tell an AI agent from a bot?
You don't have to. AbuseZero Automation Defense works inside every evaluation and AbuseZero verifies AI agents itself. You choose one of three options per benefit, and every claim comes back allow or deny.
Which AI agents count as verified?
AI agents that AbuseZero has verified. AbuseZero maintains this itself, so there's nothing for you to register or keep up to date.
Do verified AI agents bypass my rules?
No. Every claim still goes through the benefit's other protection: allow & block lists, your protection level, custom policies and eligibility. A verified AI agent's claim for someone who already had the trial is evaluated like any repeat claim.
Do I still need a CAPTCHA?
A CAPTCHA asks whether someone is human, which no longer separates good automation from bad. AbuseZero Automation Defense runs inside every evaluation with nothing for your visitors to solve, and you choose per benefit how automation and AI agents are treated.
What if an agent claims many trials?
Be ready for agent traffic
Start with Builder
Pay as you go. Choose how each benefit treats automation.
Bots and AI agents
The three options and how to set them.
See pricing
Builder, Pro, Scale and Enterprise.