How do attorney advertising rules apply to blog posts and AI content?

Treat every blog post, FAQ answer, and AI-assisted draft the firm publishes as a communication about the firm's services, because it can be read that way. ABA Model Rule 7.1's ban on false or misleading statements does not relax because a post is educational or because software wrote the first draft. State versions of the rules differ, so the firm's own state rules, applied by a lawyer who reviews the post, decide what can be published.

Most firms review ads carefully and blog posts barely at all. Yet a blog post can make the same implied promises an ad does, stay online for years, and get quoted in an AI answer far from the page it came from.

This is a content-planning question for firm owners and marketing leads, not legal advice. Whether a particular post counts as advertising under your state's rules is a question for ethics counsel.

Where the Model Rules start

Model Rule 7.1 says "A lawyer shall not make a false or misleading communication about the lawyer or the lawyer's services," and a communication is misleading if it "omits a fact necessary to make the statement considered as a whole not materially misleading." Model Rule 7.2(a) says "A lawyer may communicate information regarding the lawyer's services through any media," and 7.2(d) requires communications under that rule to include "the name and contact information of at least one lawyer or law firm responsible for its content."

The comment to Model Rule 7.3 draws a line that matters for websites: "A lawyer's communication is not a solicitation if it is directed to the general public, such as through a billboard, an Internet banner advertisement, a website or a television commercial." Under that comment, a blog post on the firm's public website, directed to the general public, is not a solicitation under the Model Rules, while the misleading-communication rule still applies to it.

These are model rules. The ABA's comparison charts "show how each jurisdiction has modified each of the ABA Model Rules of Professional Conduct," which is why every review should name the state whose rules it applies.

The 7.2(d) requirement is easy to overlook on a blog. A post with no author and no contact details may sit on a site that carries both elsewhere, but the safer habit is to make the responsible lawyer or firm obvious on every post, then confirm what your state requires.

Where blog posts tend to drift

Educational posts rarely make a false claim on purpose. They drift. Educational is not the same as exempt, and the drift usually shows up in a few predictable places:

• Implied outcomes, such as "we get clients the compensation they deserve"
• Comparisons the firm cannot support, such as "the most trusted firm in the county"
• Expertise wording, since Model Rule 7.2(c) limits stating or implying certification as a specialist
• Stale law, where a post describes a rule or deadline that has since changed
• Hypotheticals built so close to a real client's matter that the client could be recognized

Each of those is easier to catch in a review checklist than in a quick read. The last two are not advertising questions at all, which is a reason the reviewer should be a lawyer rather than only a marketer.

Stale law is the quietest risk. A post written accurately three years ago can become wrong without anyone touching it, and nothing on the page tells the reader. A review date on every legal explainer, with a named owner who keeps it, is the simplest safeguard.

What AI drafting changes

AI does not change the rules, but it adds new ways to break them. ABA Formal Opinion 512 warns that some generative AI tools are "prone to 'hallucinations,' providing ostensibly plausible responses that have no basis in fact or reality," and that "uncritical reliance on content created by a GAI tool can result in inaccurate legal advice to clients." The opinion focuses on legal work, not marketing, but a footnote flags Model Rule 7.1 as a possible additional issue.

Opinion 512 also says "Managerial lawyers must establish clear policies regarding the law firm's permissible use of GAI," and that supervision duties "also apply to GAI providers and tools." A firm that lets a vendor or a junior staffer publish AI drafts without a policy is relying on luck. None of that forbids AI drafts. It means the review has to assume the draft may contain errors that read well, which is a different posture from editing a draft written by an attorney who knows the subject.

Google's guidance on generative AI content says "It is critical to manually factcheck and review all AI-generated content for accuracy and trustworthiness before publishing." For a law firm, that check has a second layer: accuracy about the law, and accuracy about the firm.

Build review into the publishing workflow

A review step works when it has named owners and a fixed set of questions. Before any post publishes, the editor should be able to answer, "Who approved this post, and against which state's rules?" If nobody can, the post waits; a week's delay costs less than correcting a published claim.

The table works best as a literal checklist attached to each post in the content calendar. A reviewer who has to remember six questions will eventually remember five, and a post that passes the list can still be dull, but it is far less likely to embarrass the firm.

A fictional example

Consider a fictional personal injury firm in Colorado that receives an AI draft for a post about rear-end collisions. This is an invented example. One sentence in the draft reads: "Our aggressive attorneys always secure maximum compensation for every client."

The reviewing attorney flags three problems: "always" implies an outcome, "maximum" is a comparison nobody can support, and "aggressive" describes no actual practice. The revision reads: "Our attorneys review the police report, medical records, and insurance correspondence before advising a client on next steps." That sentence is checkable, the firm can confirm it describes what it actually does, and a reader learns what the firm will actually look at. The reviewer also checks the post's statements about Colorado law against primary sources, because the draft cited none.

The better fix comes earlier. If the AI had been drafting from an approved account of the firm's process, it would have had no reason to invent adjectives. Nobody, human or software, should be left to make up what the firm does; the firm supplies that, and the draft reflects it. The page on whether a law firm should use AI to write its blog covers that approach, and how to present case results covers the hardest claims to get right. The article on whether personal injury content makes sense when quoted alone adds a useful test.

In Hudson South's process, Authority Asset gathers the firm's approved knowledge and evidence, Hudson Twenty-One records how the firm communicates across 21 dimensions, Market Intelligence researches the questions clients ask, and AEO Creation turns those inputs into answer-first content for attorney review. Our law firm content guide lists the related questions.

Start from what your firm can support

On a free 30-minute call, Hudson South walks through how your published content compares with the firms prospective clients are reading.

Sources

1. Rule 7.1: Communications Concerning a Lawyer's Services. American Bar Association. Read 2026-10-15.
2. Rule 7.2: Communications Concerning a Lawyer's Services: Specific Rules. American Bar Association. Read 2026-10-15.
3. Rule 7.3 Solicitation of Clients, Comment. American Bar Association. Read 2026-10-15.
4. Jurisdictional Rules Comparison Charts. American Bar Association. Read 2026-10-15.
5. Formal Opinion 512: Generative Artificial Intelligence Tools. American Bar Association Standing Committee on Ethics and Professional Responsibility. Read 2026-10-15.
6. Google Search's guidance on using generative AI content on your website. Google Search Central. Read 2026-10-15.