Jane Smith is Field Chief Data & AI Officer at ThoughtSpot, the world’s leading Agentic Analytics platform. Jane was previously Chief Data Officer at Simply Business, a leading InsureTech and has held data leadership positions at GlaxoSmithKline and Ipsos. Jane is also a keynote speaker focusing on business adoption of AI as well as the broader societal, ethical and philosophical implications of AI.
Recently, in an exclusive interview with CIO Magazine, Jane shared insights into how her journey from Data Product Director for Manufacturing sparked her metamorphosis into a data builder passionate about shipping production systems that drive revenue, not theatre. She believes the most transformative sector in EMEA is professional services, where audit, tax and advisory workflows are being re-architected into governed enterprise intelligence platforms that will reshape the billable-hour model by 2030. Looking ahead, her goal is to build a career around consequential problems and be recognized as an authority on AI and business strategy, advising that aspiring CDAIOs embrace the chaos of innovation but know when to formalize MLOps and handovers to truly ship. The following excerpts are taken from the interview.
Hi Jane. What was the moment early in your career when you realized your path would be about turning data into production systems, not just strategy decks?
I used to be data product Director for Manufacturing at GlaxoSmithKline. It was an incredibly fun time to work in data and machine learning circa 2021, and I was leading a whole lot of build work – everything from predicting risk of individual plants, to digital twins for production lines to looking at reducing carbon output in factories. Building data products and tools that people used and benefitted from in an industry literally in the business of making lives better… that was the start of my metamorphosis into a data builder.
Your passion is clearly about making AI real — infrastructure that ships and drives revenue, not theatre. What is it about this builder’s mindset that still excites you after 20 years inside enterprises?
Hype and futurism have their place. It’s how we attract attention and funding and people into something new, but they’re not the same as building something to completion and seeing it drive a business forward. When you build, you see raw materials like data as assets and make them into something greater than what they were before. When I was growing up my father was a builder – we had piles of sand and timber etc outside our house which I was never allowed to play on because they were ‘piles of cash!’ This mindset has never left me.
Recently, you shared the story of writing your own event pipeline when your open-source collector moved to subscription and compared it to today’s “I’ll build it myself in Claude.” Why do you think so many brilliant data teams are still tempted to build their own in Claude for production, knowing they’ll end up with an unsupported custom build with no vendor to call?
Pride of ownership… pride of building… We love to be able to build ourselves and know that it’s ours. I’m as guilty as anyone in this space! It’s a pioneering spirit to say you’ll build something yourself and, in some cases, a great way to avoid a lot of cost. 20 years into this line of work I would say that being judicious about what you build and WHY is the key. Build what is genuinely your strength and differentiator, but reinventing the wheel so you feel proud about it? Not so much!
As Field CDAIO, you see many industries at once. Which industry in EMEA do you think is most quietly transforming through data and AI, and what will it look like by 2030?
Professional services are undergoing a big shift. Firms are re-architecting deep audit, tax and advisory workflows turning vast institutional knowledge into governed enterprise intelligence. In a few years the billable hour model will likely have transformed into platform and SaaS offerings that enable real-time assurance and services. I find this tremendously exciting as audit is such a fundamental part of how businesses and economies function and transact safely and reliably – we all get audited!
Talent, trust, and technology are often cited as barriers. What emerging trend in enterprise team structure or operating model do you think will solve the talent and trust bottleneck for AI?
Shifting away from centralised AI teams to smaller domain pods where the person who owns the outcome also owns the AI. The role of centralised teams is likely to focus on increasingly automated and embedded governance over review boards. The pods amplify and codify institutional knowledge. We’re already seeing this in leading companies and at many of them, analysts are leading the charge towards this change as they tend to be tightly coupled with the business anyway and ideally placed to domains build capability for this new era.
In your post, you asked fellow Chief Data / AI / Product Officers what they’ve changed their mind about. What is the one belief you held strongly as a CDO that you would now tell your younger self to let go of?
That tension between the product teams who wanted to innovate and move fast and the architects who were building for safety and security years into the future was something to be healed. I now know this is healthy tension – both teams have different incentives and objectives and the healthy tension is what you actively want to have.
Leaders who live in the messy, political reality of enterprise AI need ways to reset. What is your favorite way to disconnect and think clearly when the enterprise noise gets too loud?
I love hiking and managed to hike in Arizona this year. I underwent a full training protocol of steps to take in the event of a rattlesnake but almost disappointingly didn’t see one in the end. I go for walks a lot around where I live in the very snake-free Surry Hills but recently had to go to A&E after an injury to my foot caused by…. a twig. I know. No glory in that tale at all.
With a background in law, data, and international keynotes, your curiosity is wide-ranging. Is there a favorite book, legal case, or idea that fundamentally changed how you think about risk and responsibility?
A trip to the Bank of England Museum that I took with my teenagers last summer – they did not thank me for that, I can tell you. Learning about the Rothschild family and the beginning of the banking system in the UK made me realise that without regulation, banking would never have scaled as it was very risky – you had no guarantee you would get your money back! That made me think about risk and regulation in a new way, not as a burden but an accelerator.
What is your biggest goal? Where do you see yourself in 5 years from now?
I want to build a career around consequential problems… in five years I’d like to be recognised as an authority on AI and business strategy and influencing decisions that matter across companies, countries and economies.
You have built teams, governance, and production infrastructure from the ground up. What is the single most important advice you give to aspiring Chief Data and AI Officers who want to build teams that actually ship?
Embrace the chaos, mess and disaster of innovation, knowing that only a few genuinely amazing things will emerge. And know when the mess is over and your operation needs to grow up and ML opps, handovers between data scientists and engineers need to be implemented and formalised.