Prompt Like A Human, Not An ‘Engineer’ – Artificial Lawyer

By Antti Innanen.

Prompting is changing. We used to have elaborate prompts and ‘prompt engineers.’ And it used to be very useful. The first versions of AI tools needed quite a lot of guidance. If you had good, structured prompts, you got better results.

But things are different now. The new models are much better at understanding user intent and natural language. These models think, meaning that ‘chain-of-thought’ reasoning is already happening inside the machine. You don’t need to tell it to go step-by-step, it’s doing that by default.

And sometimes, structured prompts or bland instructions actually hurt performance.

Enter the Vibe

This shift has brought a new prompting idea into focus: vibe. The term comes from AI expert Andrej Karpathy.

Instead of laying out detailed, step-by-step instructions, ‘vibing’ is about guiding the model through mood, tone, intent, and cultural references. You tell it how something should feel instead of spelling out how it should work. You trust the model to connect the dots.

Talk Like a Human

It sounds a bit odd, but it really works. AI tools see the world differently. Instructions that seem easy for humans can feel vague to an AI. Let me give you an example:

I recently caught myself prompting with a meme. You know when AI tries too hard to sound casual or relatable? Like a cool uncle?

I told the AI to be ‘a bit less ‘how do you do, fellow kids.’’

It’s a reference to the TV series 30 Rock, where Steve Buscemi awkwardly tries to blend in with teenagers. Painfully accurate.

What’s funny is…it’s actually a great prompt. Way better than just saying ‘try to be less cool.’

Here’s how the AI processes those two prompts:

‘How do you do, fellow kids’ reference:

Specific cultural cue that evokes a clear image and feeling

Pins down the type of awkward behavior to avoid

Memorable and easy to model against

Gives a behavioral pattern the AI can recognize

‘Try to be less cool’:

Vague and ambiguous

‘Cool’ could mean anything

No clear direction for what to change

Open to multiple, conflicting interpretations

From the model’s point of view, ‘fellow kids’ is way more actionable. It’s specific, pattern-based, and full of context.

Large language models are trained on a massive, messy pile of human culture: books, conversations, memes, jokes, cringe posts, heartfelt blogs. They’re not just parsing grammar. They’re absorbing tone, subtext, and rhythm.

That’s why ‘a bit less ‘how do you do, fellow kids’’ is such a good prompt. It communicates context and intent.

Prompt Less, Converse More

So, what should you take from this?

First: drop the old idea that prompting means giving commands. That mindset leads to hallucinations and poor results. Instead, treat it like a conversation.

Second: use regular language. It carries tone, intent, and nuance. It’s more effective than blunt instructions like ‘draft me a contract.’ Talk to the tool like you would to a colleague. Ask questions. Experiment. Use 100 small prompts instead of one big one. Move between models and conversations to get better perspectives.

Even in legal contexts, a conversational style gets better results and fewer hallucinations. It keeps you engaged and thinking. Commands outsource your thinking, conversation activates it.

The hottest new prompting style is to have a conversation with the tool.

And the hottest new prompting language is… just normal language.

—

About the Author: Antti Innanen is a tech lawyer, legal design enthusiast, and an AI geek. His new book ’Prompted: How to Create and Communicate with AI’ will be published by Routledge in 2025. You can find more about Antti and his work here: Legit; Dot.: www.dot.legal and the Legal Design School.

–

[ This is an educational think piece by Antti for Artificial Lawyer. ]

—

Legal Innovators California conference, San Francisco, June 11 + 12

If you’re interested in the cutting edge of legal AI and innovation – and where we are all heading – then come along to Legal Innovators California, in San Francisco, June 11 and 12, where speakers from the leading law firms, inhouse teams, and tech companies will be sharing their insights and experiences as to what is really happening.

We already have an incredible roster of companies to hear from. This includes: &AI, Legora, Harvey, StructureFlow, Ivo, Flatiron Law Group, PointOne, Centari, LexisNexis, eBrevia, Legatics, Knowable, Draftwise, newcode.AI, Riskaway, Aracor, SimpleClosure and more.

Cooley, Wilson Sonsini, Baker McKenzie, Gunderson, Ropes & Grey, A&O Shearman and many other leading law firms will also be taking part. 

See you all there!

More information and tickets here.