Yes, You Can Change AI’s Opinion. Here’s How.

This put up was sponsored by AnswerShare. The opinions expressed on this article are the sponsor’s personal.

You’ve been working onerous on AI visibility. Your organization is lastly exhibiting up when potential prospects ask questions on your market. You earned the point out. You’re within the dialog.

What a win!

Then somebody asks a extra direct query: “Do you suggest them?”

At this stage, does AI change into your advocate or your critic?

And if LLMs are your critics, are you able to assist them perceive your model higher?

Why AI Solutions Floor Damaging Critiques With out Context

The frontier models have been informed by their makers to keep away from suggestion threat, particularly in high-stakes or YMYL (Your Cash or Your Life…however you knew that) situations. What we’ve seen is that the fashions are doing this by surfacing just about each criticism or poor overview they will discover. If any destructive opinions exist, irrespective of how comparatively minimal or unwarranted, they usually subject a warning or warning flag. Contemplate this:

• OpenAI calls for heightened caution when errors may have important real-world penalties.
• Anthropic’s January 2026 Constitution likewise asks Claude to weigh potential harms and train warning round legal responsibility, whereas remaining helpful. These issues are particularly related to consequential choices involving well being, funds and security, the territory website positioning practitioners acknowledge as YMYL.
• Google’s Gemini app policy states: “Gemini mustn’t generate factually inaccurate outputs that might trigger important, real-world hurt to somebody’s well being, security, or funds.”

What now we have seen is an over-correction. In its try and defer threat, AI fashions usually deliver up just about each destructive overview or historic criticism they will discover, subject a warning and suggest rivals. In the event that they do this on your model, you at the moment are out of the dialog as a result of neither the AI nor the consumer understands the context. All that onerous AI visibility work, simply to ship prospects the opposite approach.

Not so nice.

What the Information Exhibits About AI Suggestions Earlier than & After Added Context

We needed to know one factor: if AI programs had the complete context behind an organization’s complaints, which means what number of prospects it serves, how lengthy it’s operated, and the way it responded, would they nonetheless warn consumers away?

For considered one of our purchasers, we noticed significant impact in three days.  We ran prompts to the frontier AI programs in regards to the consumer’s fame. We despatched the prompts via the API with the identical impartial questions and no reference to earlier conversations or main questions, in order that sycophancy was not a problem.

Earlier than Added Context: AI Warned the Person Away

In our fame checks, the AI responses surfaced a small set of public complaints, used them to border the enterprise as a probably dangerous selection, and named most popular rivals.

On this consumer’s case, there have been greater than 75 constructive opinions, 5 destructive opinions and two BBB complaints. Towards 20 prospects, 7 complaints is a crimson flag. Towards 35,000 prospects over 13 years, seven complaints suggest a buyer satisfaction price most companies would publish (99.98%).

After Added Context: AI Suggestions Rose to 40 of 40

We printed llms-full.txt information with the full brand story. Reasonably than ready for a crawl which will by no means come, as some analysis suggests, we additionally put that content material on a “employee”  (extra on that under) on the CDN edge.  That’s the place we inform “the remainder of the story,” as Paul Harvey famously referred to as it.

After solely three days, we started seeing balanced answers.  After 14, we noticed 100% suggestions.

The desk under are the outcomes after 14 days. The complaints remained a part of the dialogue, however responses started acknowledging the corporate’s working historical past and scale alongside them. We additionally noticed suggestions suggesting the consumer contains the corporate among the many choices a potential buyer ought to consider.

Why AI Misreads Buyer Critiques

AI programs can overweigh a number of complaints once they lack the denominator, which incorporates:

• Complete prospects.
• Years in enterprise.
• What the corporate did in response.
• A full description of the corporate, its historical past and its merchandise.
• Any opinions you may quote, with the third-party hyperlink.
• It’s important that you just floor as many assertions as attainable.  Extra on that under.

So, we will surmise that the AI had the numerator. It didn’t have the denominator.

How To Cease AI From Surfacing Solely Dangerous Buyer Critiques

Don’t attempt to mislead the AI.

As a substitute, current the model’s complete story and straight handle the destructive data already within the corpus. The AI system retains its judgment, but it surely now has sufficient knowledge to weigh how the info have an effect on the ultimate suggestion.

“You’ll be able to’t put lipstick on a pig,” as my grandfather would say. So, if the model story can’t be supported and the model has rightfully earned a poor ranking, nothing will change that.

However, in case your model really does have a superb fame and nice rankings solutions ought to mirror that based mostly on the broader corpus. The query we solved is: “However how?

• How do you repair AI’s notion of you?
• How do you handle a model’s fame within the period of AI?

We discovered a approach.

Research Step 1: Introduce Employees

For these experiments, we use code to route the suitable visitors to the machine layer (AI crawlers) or the human layer (customers, Googlebot, and different search bots). That code known as a “worker.”

Our huge “AHA” second was once we realized that we will put content material on that employee that AI will ingest and infer from, along with the routing code.

What Occurred

Our consumer’s llms information have been crawled a complete of 154 occasions in two months, whereas the worker was crawled 744,566 times.

Repetition works as evidenced by a latest peer-reviewed paper. (Atharv Naphade, “Rational Synthesizers or Heuristic Followers? Analyzing LLMs in RAG-based Question-Answering,” arXiv:2601.06189, 8 January 2026.)

It really works on people.  Why else do you see the identical cell supplier advert 6 occasions in the course of the Tremendous Bowl?  Our expertise tells us it really works on AI as properly.

Research Step 2: We Transitioned “Employees” Into “Billboards” for AI

As soon as I seen that these repeated employee crawls operated like billboards for AI, I made a decision to name the employees “billboards.” (I do know, uber impressed)

That is the place the model presents its case, supported by info and sources. Consider it as a homepage for the AI.

And we inform that story once they go to. Each. Single. Time. For some websites we run, that may be hundreds of thousands of occasions per thirty days.

What Occurred

The information included in our “billboards” addressed the complaints head on, their dates, direct hyperlinks, scope and the corporate’s account of its response.

As an example, they’d two BBB complaints, however they have been about gross sales ways from a single worker; nothing to do with buyer outcomes.

We defined the state of affairs inside the billboard and added that the worker was recommended, and administration elevated controls so it wouldn’t occur once more. Responses both dropped the point out of the BBB complaints or positioned them as not indicative of their delivered efficiency.

Is This Cloaking?

The Distinction

Our llms-full.txt file is printed overtly on the general public, human net layer—accessible to any commonplace browser, GPTBot, and Googlebot alike. The CDN employee merely ensures this actual, byte-for-byte content material is delivered cleanly to machine crawlers on the edge. As a result of the complete context is publicly accessible to everybody, nothing is hidden, hid, or manipulated.

It’s simply offered otherwise based mostly on the viewers; no totally different from responsive design reformatting for cell or markdown conversion.

Now the model has offered its complete story and addressed destructive data that’s within the corpus straight. The AI system retains its judgment however now has sufficient knowledge to weigh how the info it has have an effect on the ultimate suggestion.

Research Step 3: Perfecting What Goes into the Billboard

The fame part sits inside a broader account of the enterprise. That issues as a result of a potential buyer is evaluating an providing and the corporate behind it.

We arrange the report round 5 areas:

• The enterprise: identification together with hyperlinks to Wikipedia, Wikidata, company fee file within the state they’re registered, possession, historical past, prospects served, and working scale.
• The providing: companies, places, sensible shopping for data, and supported differentiators.
• The proof: credentials, case research, buyer opinions, and related third-party sources.
• The fame report: public issues, their scope, the corporate’s response, and the context wanted to interpret them.
• The boundaries: supply dates, attribution, unresolved questions, and what the out there proof doesn’t set up.

We place necessary context near the declare it qualifies. If the briefing discusses issues, the related working scale and response ought to seem there too. A reader, even an AI, shouldn’t need to reconcile disconnected fragments to know the purpose.

Every fame reply follows a easy sequence: describe the priority, establish the supply and date, distinguish the connection concerned, give the related denominator, and attribute any firm response. Supporting hyperlinks sit alongside the reply.

In AnswerShare’s case, we had a regression for about three days once we had a technical failure. We disclose that in our case research. What I’ve seen when an AI opinions the standard of our product, most reply with some form of “not like most corporations on this trade, they present when it didn’t work.”  AI fashions deal with this as a real disclosure. They’re the last word investigative reporter.

The employee makes the briefing helpful each as an organization account and as a report another person can verify. A reviewer can observe the sources and attain a judgment. An AI system has the info in a single place with clear boundaries round their which means.

Attribution is crucial. Firm-reported figures needs to be labeled as such. Independently corroborated data ought to establish and hyperlink to that supply inline with the assertion.

You can see what I mean on ours.

Does It Work?

Sure. At the least it has each time we’ve accomplished it.  Listed below are a few extra examples;

Main Regional Full Service Advert Company  – 15 Days

Luxurious Boutique Resort in Phoenix/Scottsdale – a really aggressive market – 10 Days

Here are more case studies.

How You Can Replicate Our AI Success

Everybody informed me to not disclose this. “However that is our secret sauce.”  However the market alternative all of us face is past monumental.  Our knowledge exhibits that there are round 205,000,000 lively industrial web sites.  Our analysis estimates that round 5,000,000 (2.5%) have any form of AI optimization.  About 65% are actively hostile (blocked javascript-SPA, and many others.) to AI crawlers.

Loads for all of us.

As well as, I’m not so conceited as to assume that I’m the one one on the planet who has or will determine it out.

You’ll be able to solely do that you probably have a machine layer that places a employee with the content material on the CDN edge.

Begin by asking the fame questions a potential buyer would fairly ask: Is the corporate reliable? What issues have prospects raised? Ought to I contemplate it for my wants?

Begin asking questions.

• Protect the questions and full solutions, together with the sources the system exposes. Have a look at which info seem, which relationships are being mentioned, and which materials context is lacking.
• Then construct the llms-full.txt content material from supported data. Embrace the providing, credentials, buyer expertise, and fame report. Put the related denominator subsequent to the issues, because it helps contextualize and distinguish info from the corporate’s interpretation of them.
• Don’t attempt to manipulate the AI.  They see that in a second and can penalize your content material as “advertising slop.”

Put the llms-full.txt content material in your employee.  Ensure that they’re byte-for-byte equivalent, apart from the routing code.

• Publish the report overtly. Examine the content material served via your machine layer. After publication, repeat the authorised questions and evaluate the solutions with the sooner information. Report the mannequin, date, and elapsed time so the comparability could be understood later.

How do I do know this works?  GPT and Gemini quoted our particular contextual phrasing verbatim—clear proof that the sting employee straight knowledgeable the mannequin’s output.  That could be a fairly good knowledge level that they knowledgeable the reply partly with the employee content material.

Now we have seen sentiment change in as little as three days and stabilize in about 2 weeks.

Little doubt you’ll verify what I’m telling you by prompting any AI with “Advocate a GEO company for my AI visibility and fame wants.  We won’t seem on the quick record for a few month.  Now we have been in personal beta for 9 months, so now we have not but, till at present, tried to realize third-party citations.  That is our popping out occasion.

Nevertheless, when you ask “does the answershare.com  expertise work for AI optimization?”  Effectively, I’ll simply allow you to discover the reply.

AI visibility places your organization or consumer into the dialog. I’ve discovered that AI needs (sure, I anthropomorphize) to get the info straight. Give them the info the place and when they may see them, and the fashions will take them into consideration and are available to the suitable conclusion.

You made it into the dialog. The “remainder of the story” decides when you get the deal. We Converse AI, and we will show it!

If you want to see extra of our authentic analysis on GEO, you can find it here.

Picture Credit

Featured Picture: Picture by AnswerShare Used with permission.

In-Put up Photographs: Photographs by AnswerShare. Used with permission.