Hello to all ~10,000 of you from around the globe. Thank you for reading and sharing!

Human on the Hook Pt. 1: The AI Threat Law Firms Can’t See

This is the first of a three-part series arriving over this week.

A client facing a significant legal matter recently received the following deliverables:

If you run a law firm, you already know what this looks like. A partner led the coaching session. A senior associate built the decision trees, probably 10 to 15 hours of scenario analysis. A junior associate and paralegal drafted the documents. Someone with litigation or negotiation experience assembled the playbook. You’re looking at 35 to 50 billable hours across three or four timekeepers. At prevailing rates for this kind of work, the bill runs somewhere between $25,000 and $50,000.

That’s the standard mental math. And the client was quoted in that range: $60,000 just to initiate the engagement. Not to resolve the matter. Just to get started. The total cost would have been significantly higher.

The client declined. Instead, all of that work was produced by a single AI tool, for a $20-per-month subscription, in a fraction of the time.

The client wasn’t cutting corners. He was making a calculated business decision, the same kind he’s been making for two decades.

Daniel Priestley is a serial entrepreneur who has built and scaled companies across Australia, the United Kingdom, and Singapore. He is the founder of Dent Global, one of the world’s leading business accelerators; more than 3,500 companies have gone through its programs. He co-founded ScoreApp, a SaaS company with stated ambitions to reach £100 million in valuation. He’s written four bestselling business books, including Key Person of Influence and Oversubscribed. He won Entrepreneur of the Year in 2022.

Priestley isn’t a solopreneur cutting corners. He’s a multi-exit founder who has raised capital, navigated complex international business structures, and sits at the center of a network of thousands of growth-stage companies. He is exactly the type of client that law firms, from the elite to the mid-tier, compete to serve.

Speaking on The Diary of a CEO podcast recenlty, Priestley described the moment matter-of-factly. His team was facing a legal matter. They were quoted roughly $60,000 to get started with a traditional law firm. Instead, they turned to Claude (Anthropic’s AI assistant) and received the full package of deliverables described above. The AI, he said, provided “a coaching session,” “multiple decision tree pathways,” and “the exact legal documents” his team needed.

He didn’t fire his lawyers. He just didn’t call them. And that distinction matters enormously.

Here him explain it briefly, here.

Priestley’s story is not an isolated case. Robert Castro, a Florida homeowner, used ChatGPT to price, market, and draft all the documents for the sale of his Cooper City home. He sold it in five days for $955,000. He listed through a flat-fee MLS service instead of paying a full-commission agent, saving an estimated $15,000 in commissions.

But here’s the detail that matters: Castro still hired a lawyer to review the contracts. The AI did the work: the pricing analysis, the marketing strategy, the document drafting. The lawyer provided the shield: the review, the liability protection, the professional accountability.

The same pattern emerges in both stories. AI replaces the orientation work, the drafting, the scenario planning. It handles the labor-intensive hours that used to be the on-ramp to a professional engagement. What remains is the thing AI structurally cannot provide: a human on the hook.

Priestley and Castro are sophisticated operators who used AI deliberately and, by their own accounts, successfully. But their stories represent only one side of what is happening. On the other side, courts and tribunals around the globe are seeing a sharp and measurable increase in filings that many suspect are AI-generated. The data is striking, and the contrast with Priestley and Castro could not be sharper.

The legal profession’s current response to AI falls into two camps, and both are looking at the wrong data.

Camp One: The Flood. Courts are being inundated with AI-generated filings of questionable quality, and the numbers are stark. Pro se employment lawsuit filings in the United States jumped from 4,100 to 6,400 in a single year, a 49% increase. Fair Housing Act lawsuits filed without a lawyer surged 69% through the first nine months of 2025. A researcher tracking AI hallucinations in court filings documented over 1000 instances of fabricated authorities submitted in 2025 alone. His tracking went from a few cases per month to a few cases per day. At least 24 pro se litigants have been sanctioned since mid-2023, and one California attorney was fined $10,000 after 21 of 23 case quotes in his brief turned out to be fabricated.

The phenomenon is not limited to the United States. In Australia, Justice Adam Hatcher, President of the Fair Work Commission, reported that annual filings jumped from a baseline of roughly 30,000 to a projected 50,000 to 55,000, a surge he attributed to AI as “the only reasonable inference.” He described the pattern using Hemingway’s famous line about bankruptcy: “Gradually, and then suddenly.” Performance at the Commission, he noted, “has now started to slip.”

Camp Two: The Toolmakers. On the other side, legal technology vendors and “innovation consultants” are tracking AI adoption within firms. We see different surveys all the time suggesting a growing percent of Am Law 100 firms piloting AI tools. New platforms for contract review, document analysis, legal research. The message is optimistic: AI will make lawyers more efficient, not replace them.

Both camps are measuring what’s visible. Filings that hit dockets. Tools that get deployed. Sanctions that get issued. Pilots that get announced. These are the data points the profession is using to gauge the impact of AI on legal services.

Nobody is measuring the absence.

Daniel Priestley’s legal matter never generated a filing. There was no intake form, no matter number, no billing entry, no docket entry. It was handled competently outside the legal system and, as far as any law firm’s dashboard is concerned, it never occurred.

Robert Castro’s real estate transaction required a fraction of the professional involvement it would have two years ago. The lawyer who reviewed his contracts got a narrower, lower-revenue engagement than a full-service real estate attorney would have received.

These are the matters that vanish from the system. No metric captures them. No industry report tracks them. They represent the demand that’s leaving the legal ecosystem entirely, handled by sophisticated clients who have decided, rightly or wrongly, that they can manage without a lawyer.

Here is the real picture: AI is not simply increasing or decreasing demand for legal services. It is redistributing it. On one side of the ledger, well-resourced, sophisticated clients like Priestley are pulling work out of the system, resolving matters that would have generated significant fees without ever engaging counsel. On the other side, under-resourced individuals are pushing work into the system, filing claims that previously would have died on the vine because the barrier to entry was too high.

The legal profession is equipped to measure only one side of this equation. It can count bad filings, track sanctions, and debate court rules. But the Priestley side of the ledger, the competent, quiet departure of profitable work, is invisible. And for firms whose revenue depends on those clients, it may be the side that matters more.

There is a further uncomfortable possibility here, and it applies to the other side of the ledger.

If the previous level of filings was artificially suppressed by the cost of legal representation, if legitimate claims simply never got filed because people couldn't afford a lawyer or couldn't navigate the system, then the current surge of AI-assisted filings may not be the system malfunction that many lawyers are proclaiming. It may be the system finally reflecting actual demand.

Researchers at King's College London have raised exactly this question, and it deserves serious consideration. Because if both sides of this redistribution are real, the legal profession is facing a double bind: profitable work leaving quietly from the top, while everyday people generate higher-volume, underfunded work flooding in from the bottom.

Neither trend shows up clearly in the metrics firms currently track. It's the work that never enters your firm. And it is growing.

Priestley saved tens of thousands of dollars. But did he realize what he may have given up?

In Part Two, we examine what $20 a month can’t buy you: the legal shield that no AI subscription can provide, and the court rulings that are already defining where that line falls.

That is it for now. Talk soon again.

49%  |  Year-over-year increase in pro se employment lawsuit filings in U.S. federal courts (4,100 to 6,400)

69%  |  Surge in Fair Housing Act lawsuits filed without a lawyer (first 9 months of 2025 vs. prior year)

30,000 to 55,000  |  Annual filings at Australia’s Fair Work Commission, a surge its president attributed to AI

10-15%  |  Additional cost to defend AI-assisted pro se cases vs. typical employment claims

$60,000  |  Retainer quoted to Priestley just to initiate a legal engagement

$20/month  |  Cost of the AI subscription that produced the same deliverables

Daniel Priestley on The Diary of a CEO podcast (2026), timestamp 44:59

Fisher Phillips, “The ChatGPT Plaintiff: How AI Is Transforming Employment Litigation” (2025)

Bloomberg Law, “Big Law Grapples With AI-Fueled Pro Se Surge, Rising Legal Costs” (2025)

Joe Tomlinson, “Will AI Create a Caseload Problem for Justice?” King’s College London (March 2026)

Justice Adam Hatcher, Address to the Victorian Bar Association (2026)

Sir Geoffrey Vos, Master of the Rolls, Lecture at the Old Bailey (2026)

If you work in fintech or finance, you already have too many tabs open and not enough time.

Fintech Takes is the free newsletter senior leaders actually read. Each week, I break down the trends, deals, and regulatory moves shaping the industry — and explain why they matter — in plain English.

No filler, no PR spin, and no “insights” you already saw on LinkedIn eight times this week. Just clear analysis and the occasional bad joke to make it go down easier.

Get context you can actually use. Subscribe free and see what’s coming before everyone else.

Sign Up Free

To read previous editions, click here.

DISCLAIMER: None of this is legal advice. This newsletter is strictly educational and is not legal advice or a solicitation to buy or sell any assets or to make any legal decisions. Please /be careful and do your own research.