Hey friends,
Lately I haven't stopped thinking about a sentence Simon Quirke said almost in passing:
“In compliance, the data has to be right.”
It sounds obvious until you’ve lived inside the reality of regulation. Compliance isn’t like marketing where you can test and tweak in public. It isn’t even like product where you can ship a patch tomorrow. In regulated environments, being mostly right is often the same as being wrong. And the moment someone asks, “Where did you get that from?” you need an answer that stands up under pressure.
That was the thread running through my conversation with Simon, founder and CEO of Graceview. It’s also a window into where the next wave of AI winners will come from.
Simon’s path makes sense in hindsight. He started as a transactional lawyer, doing the classic six or seven day grind. He learned fast, but worked out early that private practice wasn’t where he wanted to spend his life.
So he moved into corporate roles. Head of Legal. Company Secretary. Then London, where he became VP Legal at Getty Images during a chaotic, exciting period. They were rolling up smaller businesses, digitising an entire industry, and learning how to grow at speed.
That chapter mattered, because it gave Simon a front row seat to what it feels like when a company genuinely reshapes a market. He built teams from scratch. He saw what “growth” looks like when it isn’t a slide deck, it’s late nights, messy systems, and constant reinvention.
After returning to Australia and doing the GC and COO chapter at a listed company, he hit a fork in the road.
Take the stable path inside a bigger corporate machine, or build. He chose build.
Since then he’s started multiple ventures, and what stood out to me wasn’t the list of companies. It was the mindset. The satisfaction of starting with an idea and watching it become something clients rely on, investors support, and employees want to be part of.
That “building” itch is real. If you’ve ever felt it, you know it doesn’t go away by choosing the safe option. Graceview’s mission is intentionally ambitious.
“Our mission is to make compliance effortless.”
Simon admitted it’s a lofty ambition, and that’s the point. You don’t pick an easy mission if you’re trying to solve something that matters.
They started by asking a sharper question. Where is compliance the most painful?
They landed on one area that almost every legal and compliance team struggles with. Keeping up with regulatory change, globally, across industries, across jurisdictions, without missing anything.
This is The Part Most People Underestimate
It’s not just that there’s a lot of information. It’s that it’s a moving target. Different regulators, different publishing formats, old websites, newsletters, pages that quietly update, enforcement actions, commentary, signals and warnings. And you don’t get points for being close.
If you miss something, you lose trust. And the moment trust is gone, your system can’t be the single source of truth.
The Most Useful Frame from the Conversation
Simon described how many large language model tools work today as what he called opportunistic search.
They go out to the web, pull a handful of sources, and produce an answer. The next time you ask, you might get a different set of sources. Or the “top sources” shift. Or the information changes. Or the model confidently fills gaps.
That’s fine for casual research. In compliance, it’s dangerous.
Graceview’s approach is closer to what he called determinative search.
It starts by mapping the sources you must monitor for your industry and jurisdictions. Then those sources are monitored every 30 minutes. When something changes, the system detects it and builds what Graceview calls a regulatory fingerprint.
The goal is simple to say and hard to execute. Gap free coverage with no noise. And the key here is audit-ability.
When someone asks, “Where did you get that from?” you don’t want a vibe. You want a straight line explanation. This source. This change. This time. This evidence.
Why this matters more now than it did a year ago? Simon shared a line from one of the leading law firms they spoke to recently.
“The competition used to be other lawyers. Now the competition is our clients using GPT.”
That’s the moment the ground moved. Law is the business of words. Large language models use words. Join the dots.
Clients will try to self serve more. And they’ll do it even when it’s risky, because it’s fast, confident, and feels good. That’s exactly why reliability and traceability become the real differentiators.
Data
It also points to the deeper truth underneath most AI products. Data becomes the moat. Not “data” as in what you posted online.
Data as in a curated, structured, verified pool of sources, plus the systems that keep it current, plus the domain understanding that decides what matters and what doesn’t, plus the consistency that makes workflows trustworthy.
Not a wrapper. A foundation. We talked about “wrapper apps,” products that are basically a workflow around an API someone else controls. There are brilliant founders building in that space. But Simon made a point that every founder should sit with.
If you’re building something a foundation model company can replicate just by extending their product line, you’re living on borrowed time.
Graceview’s defensibility is the hardest thing to build. The data, and the mechanism that collects it, cleans it, monitors it, and makes it auditable.
It’s not easily disreputable because it’s not easily replicable.
The “so what” layer. Graceview started with horizon scanning. What’s changing, where, and when. Now they’re moving into the next layer. So what?
What policies need to change because of this new regulation? What obligations and controls need updating? What should we draft, flag, or route to a human for review?
This is where AI becomes useful without becoming reckless. Not replacing judgement, but removing the grunt work so legal and compliance teams can become strategic partners inside the business.
And once you stop spending your life chasing updates, something interesting happens. You can finally ask, “Is our policy even good?”
A Founder Lesson Worth Repeating
Simon gave advice that was simple and sharp. If you’re not technical but you want to build a serious product, you need a technical co founder. Not just someone you outsource to. Not a dev shop. Not a contractor who checks in. A true co conspirator.
Someone in the trenches with you, bouncing ideas on a Saturday night, seeing the shape of the problem the way you do, and helping you make decisions at the speed startups require.
And then the other truth. You have to enjoy the process, including the hard bits. Because the lows can be brutally low. The highs are real, but never guaranteed. The only sustainable fuel is being rewarded by the work itself, not just the outcome.
Key Takeaway
The takeaway I’m sitting with. AI is going to make it easier to produce words, drafts, summaries, and answers. But in high stakes environments, the real advantage won’t be who can generate the fastest response. It will be who can prove the response is grounded in reality.
In compliance, trust isn’t a brand value. It’s the product.
- Aamir
📲 Resources & Links
🎧 Listen to the Podcast Episode 1 on: Spotify | Apple Podcasts | YouTube
📘 Book: The CEO Who Mocked AI (Until It Made Him Millions) by Aamir Qutub