Not on its own. An AI tool can draft a short, generic reply for an attorney to approve, but it should not post replies automatically, because ABA Formal Opinion 496 limits what a lawyer may say in response to an online review, and an AI writer does not know which detail is confidential. In GoHighLevel the safe setting is a draft that waits for approval. The bottom line: a reply is the firm's public statement, and one wrong sentence can breach Rule 1.6.
What Opinion 496 says
The opinion, released January 13, 2021, says the main ethical concern in responding to online criticism is maintaining client confidentiality. It concludes that a negative online review, because of its informal nature, is not a controversy between the lawyer and the client within Rule 1.6(b)(5), so it does not allow disclosure of confidential information, and that even if it did, a public response is not reasonably necessary. It suggests a lawyer consider not responding, or, if a client or former client is the source, say only that professional obligations prevent responding as the lawyer would like. It points to In re Tsamis, where a lawyer received a public reprimand for disclosing client information in response to a client's negative reviews.
Why AI raises the risk
An AI reply tool writes from the review text, and reviews often describe the matter. A generated sentence like the outcome was the best available given the facts confirms details and may reveal information relating to the representation. One vendor guide warns that even saying events are not accurately portrayed could reveal the lawyer was involved. Some ethics commentators caution that even acknowledging a reviewer as a client can reveal information, so a firm can reasonably choose not to reply to reviews by clients at all.
A third party guide describes GoHighLevel's Reviews AI as drafting or posting replies to Google reviews. GoHighLevel's Conversation AI has Off, Suggestive, and Autopilot modes, and whether Reviews AI offers a draft only setting should be confirmed in your account.
| Review type | Safe reply |
|---|---|
| Positive review from a client | A short generic thanks, or no reply, per firm policy |
| Negative review from a client | No reply, or the one line about professional obligations |
| Review from a non client | Flag or report it to Google, and reply only in general terms |
| Review that names the matter or staff | Do not reply, and consult the attorney responsible |
Build it in GoHighLevel
- Decide the firm's written policy first: who approves replies and which types get none.
- If the account has Reviews AI, set it to draft only, and turn off automatic posting.
- Give the AI a template with the generic lines from the table, and tell it never to mention the matter, the outcome, or the person by name.
- Create a task for the managing attorney whenever a new review arrives, and let only that person post.
- Add a rule that reviews of 3 stars or below are routed to a person and never answered by AI.
- Log every review and reply in a simple record, and review the policy every quarter.
Worked example
For example, if a firm receives 6 new reviews a month and its policy is to reply only to positive reviews with a generic thanks, that is about 4 short replies (6 times 0.7) for an attorney to approve. It takes minutes, and it avoids a bad reply to the other 2.
Mistakes to avoid
- Autopilot posting. Draft only.
- Letting AI respond to a negative review. Opinion 496 advises against detail.
- Replying by name. That can confirm representation.
- No policy. Staff will improvise.
How this was handled before
Partners answered reviews themselves, or ignored them on advice. Automated replies came from marketing tools built for restaurants and stores, where a friendly reply is always safe. For lawyers, the ABA opinion makes the reply itself a risk, so the tool becomes a drafting aid at most.
What to measure after launch
Track reviews received, drafts approved, replies posted, and replies edited by an attorney. If attorneys rewrite most drafts, tighten the template or stop using AI for replies.
Check before you switch it on
US text messages sent from a standard 10 digit number need A2P 10DLC registration. The HighLevel support portal says registration is required for texts to US recipients from 10 digit long code numbers and that toll free numbers do not require it. HighLevel's opt in guidelines also say a person cannot be forced to agree to text messages in order to submit a form, so keep the consent box optional. One compliance guide separates informational texts, which need documented consent, from marketing texts, which need prior express written consent. Ask your attorney which category your reminders fall into. Confirm the feature and its controls in your account before relying on them, and check your state's rules on responding to reviews. This is general information, not legal advice.
Questions people ask
Can a lawyer reveal facts to defend against a negative review?
ABA Formal Opinion 496 says a negative online review does not trigger the self defense exception, and a public response would go beyond what is reasonably necessary.
What does the opinion suggest saying?
Consider not responding, or say only that professional obligations prevent responding as the lawyer would like.
Should AI post replies automatically?
No. Use drafts that an attorney approves.
Ready to try it yourself? Start a GoHighLevel account here.
You can also see this in action in our GoHighLevel capabilities demo.