AI Can Read the Law, But Can It Read the Room?

Banner artwork by PeopleImages.com - Yuri A / Shutterstock.com

The compliance conundrum – AI’s first look

To explore the modern interplay of AI and the work of in-house counsel, we’ll look at an imagined scenario that may feel familiar to many.

“Can you take a look at this new executive order on import tariffs?” Gilbert Cho asks, placing a marked-up printout on the desk with the calm assurance of a seasoned general counsel. “We need to understand the compliance implications and outline a practical implementation plan quickly. You’re closer to the business than outside counsel — and frankly, I trust your judgment.”

Catherine Lin, a mid-level counsel on the legal team with a background in regulatory compliance, gives a small nod. “I’ll get started right away.”

Back at her desk, Catherine opens the AI research platform the team had recently begun using for first-pass legal analysis and asks it to summarize the executive order, outline legal compliance requirements, and draft an implementation plan for a US-based electronics company.

The system returns a detailed breakdown: key provisions of the order, current legal frameworks, relevant citations, and even a preliminary set of recommendations. Catherine reviews the sources — mostly government releases, recent legal updates, and secondary analysis — and verifies the AI’s interpretation against her own notes.

When she returns to Gilbert’s office, outline in hand, he raises an eyebrow. “That was fast.”

Catherine offers a measured smile. “The AI gave background and direction for my research and analysis. I’ve checked the sources and added a few adjustments where the recommendations could go deeper.”

Gilbert scans the document. “Good. Let’s walk through the plan together.”

From first look to actually getting things done

The legal framework is solid. The recommendations aligns with the order.

Catherine leans over the printout. “The AI caught the core compliance obligations, and I’ve fact-checked the sources — it’s all grounded in current law. But it’s still pretty generic.”

Gilbert nods. “That’s the thing — it knows the law. It doesn’t know us.”

They both understood what that meant. The company’s global supply chain, its high-exposure SKUs, the particular dynamics between Finance, Ops, and Sales — none of that was in the model. And while the AI had offered a plausible implementation plan, it lacked the nuance to navigate organizational behavior.

Catherine points to a bullet point suggesting procurement process updates. “This assumes Procurement will pivot just because we say so. But unless we tie this to existing vendor contract reviews or build in some operational incentives, it won’t stick.”

“Right,” Gilbert says. “And Sales is already on edge with Q3 projections. If we roll this out without accounting for how they communicate price adjustments, we’re going to create confusion — or worse, misstatements.”

They work through the plan line by line, layering in institutional context, flagging assumptions, identifying potential resistance points, and stress-testing each recommendation against the company’s risk profile. Gilbert notes, “We need to think about where we’re taking risk — and make sure we’re aligned on where the liability lands.”

The in-house counsel skills AI can’t replace

This exchange underscores that there are core business skills that AI can’t replicate — at least, not yet. AI can parse statutes, draft memos, and even produce a halfway decent cease-and-desist. But can it persuade a VP of Sales to stop importing goods that just became taxable? Can it rewrite a policy that Ops will actually follow? Can it sit in a boardroom, read the room, and pitch a compliance strategy that earns buy-in — not eye rolls?

Any seasoned in-house counsel will tell you: The law is just the beginning. The real value lies in a set of distinctly irreplaceable human skills:

1. Trust

We’ve all seen the stories: hallucinated case law, fabricated citations. Even if AI can do the basic work, you’re still on the hook. The company expects work and recommendations that carry your verification — and your name. When the stakes are high, trust isn’t just nice to have. It’s non-negotiable.

2. Organizational fluency

Knowing the law is essential. But knowing how your company works is what makes it actionable. Who are the decision-makers? Who are the potential blockers? How does a directive move from idea to implementation? AI does not know the politics, the personalities, or past context. You do.

Any seasoned in-house counsel will tell you: The law is just the beginning.

3. Influence without authority

In-house counsel rarely have ultimate authority. You influence through relationships, timing, and tact. Convincing Marketing to drop a risky tagline or nudging Engineering to build in a privacy feature often takes diplomacy, not doctrine. AI cannot raise its hand at just the right moment.

4. Strategic judgment

AI optimizes for rules. Humans optimize for context. Should you push back on that indemnity clause or assume some risk to close the deal? Raise a red flag now, or monitor closely? These are judgment calls — shaped by nuance, experience, and a view of the bigger picture. No algorithm can replace that.

5. Crisis navigation

When things go sideways — and they will — you cannot turn to an AI model. You need someone who understands internal politics, external pressure and how to keep a tense situation from spiraling into a headline. Crisis response demands more than knowledge. It demands presence. It demands accountability.


By the end of Gilbert’s and Catherine’s meeting, the AI’s framework had evolved into something far more valuable: a compliance plan tailored not just to the law, but to the business. A strategy created by in-house counsel with the judgment, persuasion and business instincts that AI simply cannot replace.

Further reading:

ACC Resources:

Disclaimer: The information in any resource in this website should not be construed as legal advice or as a legal opinion on specific facts, and should not be considered representing the views of its authors, its sponsors, and/or ACC. These resources are not intended as a definitive statement on the subject addressed. Rather, they are intended to serve as a tool providing practical guidance and references for the busy in-house practitioner and other readers.

 Generate AI Summary
 ACC AI Summarizer can make mistakes, so double-check the results
Thank you for your feedback!