top of page
Search

Kirkland & Ellis Just Bet $500M That AI Should Be Built, Not Bought — The Lesson for Every NZ Business Owner

In short: In late May 2026, Kirkland & Ellis — the world's highest-grossing law firm, the first to pass $10 billion in revenue — announced it is investing $500 million to build its own proprietary AI platform rather than license an off-the-shelf tool. That is roughly 1% of its annual revenue, with $100 million committed this year alone. The strategic thesis behind the bet is the single most important AI lesson for NZ business owners in 2026: if every competitor can license the same AI, the AI stops being an advantage. The advantage comes from building AI into how the business actually works — its data, its workflows, its judgement. You cannot spend $500 million. But the principle scales all the way down to a $1m Auckland business, and acting on it is the difference between an AI subscription and an AI operating model.

Key takeaways

  • The news: Kirkland & Ellis is investing $500 million (about 1% of revenue, $100m this year) to build a proprietary AI platform rather than buy one.

  • The thesis: "Built, not bought" — if everyone licenses the same AI, it is no longer a competitive advantage. Advantage comes from integration into your own data and workflows.

  • Why it matters for NZ SMEs: The principle scales down. You cannot out-spend the market, but you can out-integrate your local competitors.

  • What it is not: It is not a signal to build a custom AI platform. For a $500k-$50m NZ business, "built not bought" means proficiency + workflow integration, not a $500m engineering project.

  • The action: Move from AI-as-a-subscription to AI-as-an-operating-model — Level 3-4 of the proficiency framework.

What Kirkland & Ellis actually announced

According to reporting from Bloomberg Law and The American Lawyer in late May 2026, Kirkland & Ellis is committing $500 million over the next several years — beginning with more than $100 million in 2026 — to build a proprietary generative AI platform tailored to its own legal work. The build involves around 250 of the firm's lawyers (including 100 partners) and 180 technology professionals mapping real transactional and disputes workflows into the system. External technology companies are helping build the architecture but are contractually barred from reselling it to rival firms. (Source: Bloomberg Law).

For context: Kirkland recently became the first law firm to pass $10 billion in annual revenue, reporting roughly $10.6 billion with profits per equity partner around $11.1 million. The $500 million AI commitment sits at about 1% of revenue — large in absolute terms, modest as a share of the business.

The thesis: built, not bought

The reason the Kirkland bet matters is not the dollar figure. It is the strategic logic. Most of the legal market is racing to license AI from the same handful of vendors — signing deals with the major legal-AI providers and frontier model companies. Kirkland reached a blunt conclusion: if every competitor can buy the same tool, the tool is not an advantage. It is table stakes.

So Kirkland is spending its way toward control — control of the data, the workflows, the risk envelope. The pitch to clients is that its AI is not just faster, but safer and more accountable, because the firm owns the whole stack rather than renting it.

That is the lesson. The competitive advantage in AI is not access to the model. Everyone has access to Claude, ChatGPT, and Gemini. The advantage is in how deeply the AI is integrated into the specific data, workflows, and judgement of your business — the part no competitor can copy by signing the same vendor contract.

Why this scales down to an Auckland SME

A $1m-$50m Auckland business owner reads "$500 million" and reasonably concludes this has nothing to do with them. That is the wrong read. The dollar figure is enterprise-scale; the principle is universal.

You cannot out-spend the market on AI. But you do not need to. Your competitors are not Kirkland & Ellis — they are other Auckland businesses your size, most of whom are at Level 1 of AI proficiency (occasional use of an off-the-shelf assistant). You do not need a $500 million platform to beat them. You need to integrate AI into your actual operating model while they are still using it to draft the occasional email.

"Built, not bought" for an Auckland SME does not mean building software. It means building proficiency and integration: connecting an AI assistant to your real business systems, designing the recurring workflows that matter to your specific business, and developing the capability across your team. That is something a $1m business can do in months, not years, and for thousands of dollars, not millions.

Where this sits on the Five Levels of Claude Proficiency

Kirkland is going to Level 5 of the Strategize Auckland Five Levels of Claude Proficiency framework — AI embedded into the operating system of the business, with processes redesigned around what AI can do directly. The $500 million is what Level 5 looks like at the scale of a $10 billion firm.

An Auckland SME does not need Level 5. The competitive gap in the Auckland market opens up between Level 1 (where most owners are) and Level 4 (AI running autonomous work against the business systems). Closing that gap is achievable, affordable, and is exactly the move Kirkland is making — just at a scale that fits a $1m-$50m business.

What an NZ business owner should actually do

  • Stop thinking of AI as a subscription. A ChatGPT or Claude licence is access, not advantage. Access is table stakes now.

  • Pick the workflows that matter to your business. Not "AI across the business" — the three to five recurring workflows where AI integration produces the largest measurable improvement.

  • Integrate, do not just use. Connect the AI to your real systems and data where it is safe to do so. Integration is the part competitors cannot copy.

  • Develop the capability across your team. Owner-only proficiency is bounded. Team-level proficiency compounds.

  • Move before your competitors do. The Kirkland move signals the window is now. The advantage goes to the businesses that integrate first.

How Strategize Auckland helps

Strategize Auckland runs AI strategy for Auckland SMEs — the same "built, not bought" principle Kirkland is applying, scaled to a $500k-$50m business. The practice operates at Level 4 of its own framework: agentic AI pipelines running against real business systems daily. The Claude Proficiency Advisory and the broader 52-week guidance programme (NZD $12,000-$24,000 per year, RBP co-funding for the first three months for qualifying businesses) develop the same operating-model advantage Kirkland is buying — at a scale that fits your business. Read how the engagement works.

Frequently asked questions

Does this mean my business should build its own AI?

Almost certainly not in the software sense. "Built, not bought" at SME scale means building proficiency and workflow integration on top of an existing frontier model (Claude, ChatGPT, Gemini) — not building a model or platform. The build is in how you integrate, not in engineering.

How much would the SME version of this cost?

Orders of magnitude less than Kirkland. The strategic advisory runs through the Strategize Auckland 52-week programme at NZD $12,000-$24,000 per year. AI tool licences are tens of dollars per user per month. The integration work depends on the systems involved. The total is in the thousands, not the millions.

Is this just hype because a big firm spent a lot?

The dollar figure is newsworthy; the principle is not hype. The observation that licensed AI is table stakes rather than advantage is structurally true at every business size. Kirkland is the proof point that the most commercially successful operators are acting on it.

What if my competitors are not using AI at all yet?

Then the opportunity is larger, not smaller. The first business in an Auckland niche to move from Level 1 to Level 4 opens a gap that is hard for slower competitors to close.

Want to build the "built, not bought" advantage in your Auckland business? A 15-minute call with Steven is the starting point. Book at strategizeauckland.info/book-online or call 027 737 2858.

Written by Steven Parker, Principal, Strategize Auckland. The practice operates at Level 4 of its own Claude Proficiency framework. Level 1, 55 Corinthian Drive, Albany 0632. RBP-accredited. Reviewed 30 May 2026. Kirkland & Ellis figures per Bloomberg Law / The American Lawyer reporting, May 2026.

 
 
 

Recent Posts

See All

Comments


bottom of page