The Trust-Signal Checklist for AI Visibility

You can get a first read on your Reputation, the ANSWER category that measures whether AI reflects your standing accurately, by checking two things: whether the trust signals an engine reads are strong, and what AI actually says when asked. This checklist covers both. For the full background, see the Reputation pillar guide.

Part 1: Are Your Trust Signals Strong?

Engines judge trust from what they can read. Work through these signal checks on your own footprint.

1. Recent reviews. You have a steady flow of current, genuine reviews on the platforms that matter for your category, not just a few old ones.
2. Consistent company facts. Your status, leadership, and basic details are current and consistent across your site, profiles, and major listings.
3. Visible trust proof. Security, compliance, certifications, and reliability claims are stated plainly on indexable pages.
4. Resolved complaints answered. Past issues show a public response and resolution, not just the original complaint.
5. Current coverage. Recent, credible third-party mentions reflect who you are now, outweighing old stories.
6. Clear, unique identity. Your name and brand are distinct enough that a similarly named company's problems will not attach to you.

Any "no" is a place an engine is likely to lean on an old narrative or guess.

Part 2: What Does AI Actually Say?

Now test the output. Run these prompts, replacing the brackets with your details.

1. "Is [brand] reliable and reputable?"
2. "Is [brand] a trustworthy company to buy from?"
3. "What do reviews say about [brand]'s support?"
4. "Is [brand] still in business?"
5. "Has [brand] had any security or trust issues?"

The 3 Rules

1. Log out. Run every prompt in a logged-out or private session, so the engine is not drawing on your history.
2. One fresh chat per prompt. Start a new conversation each time, so one answer does not color the next.
3. Use your real market. Run the prompts in the geography you actually sell into, because answers vary by location.

Run each on at least 2 engines, for example ChatGPT and Perplexity, plus Google's AI answers if you can.

How to Read the Results

For each answer, note whether the verdict is fair, whether a specific complaint is repeated, and whether anything is outright false. A fair, positive read means strong Reputation. A repeated old complaint points to a recency gap from Part 1. Any falsehood, a claimed shutdown, breach, or lost certification, is the most urgent finding, because it is both the most damaging and the most fixable. The worst patterns are explained in When AI Says You've Shut Down.

What This Check Cannot Tell You

A trust spot check is a useful hint, not a baseline. It covers a few prompts, on one occasion, scored by eye, and answers vary from ask to ask. A real measurement runs many trust questions, repeats them, covers more engines, and scores every answer the same way, reporting an Accuracy Rate you can track over time. Because reputation errors are the ones customers least often report, ongoing measurement matters more here than anywhere. Doing it by hand is a real undertaking, which is the honest reason most teams have it run for them.

If your check turns up false claims or stale negatives, the next step is a proper baseline and a fix plan. A HiBot audit measures your Reputation across engines and ranks what to fix first. The tactics are in How to Improve Your Reputation Score. Download the ANSWER whitepaper to see the method, or request an AI visibility audit at hibot.com.