Introduction

The questions clients bring me have changed noticeably over the past two years. The questions now include such topics as the use of AI to draft marketing copy, or to screen job applicants. Often, they are already doing all of these things — and the legal questions arrive afterward, if at all.

That order of operations is understandable, but it is backwards. The legal risks often lurk in a vendor agreement accepted with a click, in an employee’s well-intentioned shortcut, or in AI-generated work the business assumes it owns. By the time one of them surfaces, the inexpensive moment to address it has usually passed.

The Benefits of AI for Small Businesses

As I have written previously, there are benefits to AI. It can produce first drafts faster than any person, and reduce costs in ways that matter a great deal to a small business. Used thoughtfully, AI can provide genuine competitive advantages. The point is simply that it should be adopted the way a well-run business adopts anything consequential — with its contracts read, its policies written, and its responsibilities understood.

The Legal Risks of Using AI in Your Business

In my experience, the legal risks of AI for a small business fall into seven categories.

1. The Vendor Contract You Clicked Through

Every AI tool comes with terms of service, and those terms answer questions most owners never think to ask. Can the vendor use your data to train its models? Who owns the outputs the tool generates? What happens to your information if you cancel? In many standard agreements, the answers favor the vendor on every point. A contract that will touch meaningful business data deserves the same review you would give any other significant vendor relationship.

2. Confidential Information and Trade Secrets

Trade secret protection depends above all on taking reasonable steps to keep your most valuable assets secret. Pasting proprietary pricing, customer lists, or a client’s confidential documents into AI can undermine that protection — and may breach confidentiality obligations you owe under existing contracts.

3. Ownership of AI-Generated Work

If your business uses AI to create logos, marketing content, or product designs, ownership is less settled than most owners assume. The U.S. Copyright Office has taken the position that purely AI-generated material, with no meaningful human authorship, is not eligible for copyright protection. Content your business cannot protect is content a competitor may be free to copy.

4. Employee Use Without an AI Policy

Clients have often communicated that employees use AI tools at work whether or not the client has authorized them. That may mean the loss of confidential information. This suggests the necessity of an AI use policy. Such a policy would name the approved AI tools; define what information is off-limits, and require human review of anything AI-assisted before it reaches the public.

5. AI in Hiring and Employment Decisions

AI tools that screen résumés or rank applicants carry a specific exposure: if the tool produces discriminatory results, the employer — not the software vendor — generally bears the responsibility. Federal enforcement agencies have made clear that existing anti-discrimination laws apply fully to automated decision-making, and a growing number of states and localities have added requirements of their own.

6. Marketing Claims About AI

If your business advertises AI-powered products or services, those claims are subject to the same truth-in-advertising rules as every other marketing statement. The Federal Trade Commission has been explicit that it will pursue companies that overstate what their AI can do or use AI as a gloss for claims they cannot substantiate. The rule is old-fashioned: say what the product actually does, and be able to back it up.

7. Accuracy — and Who Answers for AI’s Mistakes

AI produces output that is occasionally wrong. When an AI-drafted contract term, financial summary, or customer communication contains an error, the business that sent it — not the tool that drafted it — answers for the consequences. The best safeguard is for a person with appropriate knowledge to review AI-produced work before it is used.

Conclusion: Three Steps to Get Ahead of AI Risk

A business owner does not need to become an expert in artificial intelligence to manage the risks it presents. For most small businesses, managing those risks amounts to three manageable steps: reviewing the vendor agreements regarding the AI tools you use; adopting an employee AI use policy in place, and updating your client and contractor agreements to address AI-generated work.

If your business is adopting AI tools, or already has, The Bloom Group, LLC can help you manage the risks they present. Contact us at (202) 494-3954 or peter@thebloomgroupllc.com. to schedule a consultation.

This article is for general informational purposes only and does not constitute legal advice. Reading it does not create an attorney-client relationship.