First Drafts | Why It's Unethical to NOT use AI

There’s no escaping the conversation around artificial intelligence at law firms today. Whether at solo practices or big law boardrooms, attorneys are eyeing AI with skepticism—often loudly insisting they’ll “never” use AI in legal writing. It’s an understandable sentiment, born from headlines about AI “hallucinations” and concerns about accuracy, particularly with citations. Unchecked, generative AI can produce flawed output, even with today’s advanced legal-specific models.

But let’s be frank: outright refusal to engage with AI is not just shortsighted – it’s increasingly out of step with the core duty of every lawyer under the ABA Model Rules of Professional Conduct, starting with Rule 1.1: the duty of competence.

Redefining Competence in the AI Era

ABA Rule 1.1 states: “A lawyer shall provide competent representation to a client. Competent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.”

In 2012, Comment [8] was added (and adopted by most states),clarifying that competence includes staying abreast of “the benefits and risks associated with relevant technology.”

Competence isn’t just about knowing the law or following precedent – it’s about employing the tools and processes that allow for the most thorough, well-prepared, and cost-effective representation. Today, this means understanding how AI fits into legal practice, and, where appropriate, using it.

The Real Problem: Billing Without AI Is Archaic—and Unethical

Let’s be clear: the legal profession is one of the last major industries still using essentially hand-crafted processes for document generation and legal drafting. In an environment where opposing counsel may be using AI-based tools to draft complex motions in minutes—while you’re charging your client for 25 hours of manual labor—are you truly being a competent, ethical advocate?

Imagine two scenarios:

Scenario A: You prep a 20-page motion by hand, bill 30 hours, and risk out-of-date citations or missed arguments simply because you ran out of time.

Scenario B: You use a purpose-built AI platform (like First Drafts), get a first draft of the same 20-page motion in minutes with appropriate case context, then polish with some legal research and finalize for about 5 hours of time instead.

Which approach better serves your client’s interests? Which better meets your ethical obligations of competence and diligence under Rule 1.1? Do you really want to wait until opposing counsel objects to your time entries on a motion for attorneys’ fees to find out that everyone else has been billing at an 80% discount compared to your manual drafting?

Why The ‘AI is Too Risky’ Argument Falls Flat

Let’s acknowledge reality: AI is not perfect, and it never will be. But neither are associates, Westlaw searches, or even your own human memory. Some risks are present in every form of legal research or drafting—but the answer isn’t to ignore new technology; it’s to supervise it appropriately.

The duty under Rule 1.1 is not to guarantee perfection, but to use the tools at your disposal in a way that best serves the client. The burden is on you to review and verify the work product—whether it’s from an AI tool, a junior attorney, or your own dictation system.

Transparency and Supervision: The Way Forward

At First Drafts, we’ve designed our product precisely to navigate these ethical concerns:

Only licensed attorneys can use our platform, eliminating UPL risks.
Attorneys retain full control—AI generates the initial draft, but you review, edit, and finalize.
Our structured, step-by-step approach focuses on quality and context, not “AI magic.”

The New Ethical Baseline

Competence in 2025 means knowing when to leverage AI and when to rely on traditional skills. By stubbornly refusing to consider AI, attorneys risk falling below the minimum standard of technological competence expected by the ABA, their state bars, and—most importantly—their clients.

Is it really ethical to charge full freight for manual work that can be performed 5-10x faster (and at a fraction of the cost) with the aid of carefully curated AI? When opposing counsel can pass those savings to their clients, do you really want to be in front of a judge or disciplinary board explaining why you didn’t use obvious, available efficiencies?

Conclusion