James Bergin on data & AI in 2024 and beyond
"I am hopeful that as we continue to ask these big questions about who we are, why we do what we do, and how we best work with each other, that AI could really be used for good as part of the answer."
James describes himself as ‘an architect who has never designed a building’. In the somewhat insular world of technology, that headline speaks volumes about the way he orientates and thinks about the world. I worked really closely with James during our shared four years at Xero and his ability to pull out a framework, whip up a genuinely informative metaphor and test the clarity and logic of a still forming idea with gentleness and rigour at the same time is unrivalled.
James, 2024 has been another busy year for data & AI. What’s one development / milestone / news story that really caught your eye?
The announcement of the Apple Intelligence feature-set caught my eye - particularly because I think the mega-mobile platforms are well poised to be an important default portal through which many daily interactions with AI are managed. This move by Apple, and similar moves by Google for Android, really do make me think about the ‘chief of staff’ role they could play as we start to build a team of virtual assistants that help us with multiple aspects of our lives and businesses - but a team that needs coordination in order to be effective.
I’ve also been really watching how we move towards actually creating value for small businesses out of Gen AI. I know it’s may seem bit self-serving, but at Xero getting our small business companion Just ask Xero (JAX) into beta was also a pretty big milestone for us. It has been exciting to see how generative AI tech is really enabling conversational interfaces and that, in turn, is allowing us to go where the work is taking place. So, if an electrician is conversing with their client in email, why not stay in email and forward the thread to JAX to draft the invoice? Or, if chatting with a customer in WhatsApp, chat to JAX there too. I know the teams have been working so hard on JAX, and I was really impressed to see the progress this year.
You’ve been working in and around technology, data & AI for a while now. Many things have changed! But tell us about something that was true when you started out in this space and is still important today.
That technology, at its core, is just the application of scientific knowledge for practical purposes - that’s what the word means, after all. And that when we look at the advances in knowledge and capabilities in the data/AI/other interesting parts of the tech landscape, it is crucial that we try to look at things beyond the hype and, instead, through the lens of how they can help. What problems can be better addressed by the capabilities that are being unlocked? What trade-offs am I going to have to make in order to use it.
It’s been a heady couple of years with 2024 almost as frothy as 2023. What's one common misconception about AI that you wish would go away?
Just because AI can string together sentences and create derivative artwork, does not mean that it is either sentient or capable of replacing humans outright. Many comments borne out of fear or uncertainty about the possible disruption to business and industry from “AI” seem to cite as evidence examples of specific abilities in specific contexts. I just see further underscoring of Arthur C Clarke’s famous quote - that for many people, “any sufficiently advanced technology is indistinguishable from magic”.
The festive season is almost upon us, so many readers will have a bit of extra time to read / learn / reflect. Who do you follow to stay up to date with what’s changing in the world of data & AI?
Well, you and Data Runs Deep, of course. 😉 In addition to that, though, I try to read as widely and diversely as I can, since my role requires me to look across practical, emerging and advanced technologies to look for the possible points of confluence that might lead to interesting possible scenarios of the future. I’m always open to new sources of insight and thinking and diverse points of view.
Leaning into your dystopian side for a moment, what’s your biggest fear for/with/from AI in 2025?
I recently was asked this question when speaking on a panel, and the answer that struck me then still holds now: sleepwalking. I’m worried that we just find ourselves playing with tech for tech’s sake and putting it everywhere we can without stopping to ask some of the bigger questions - particularly around where its use genuinely adds value and where it does not. So, we “sleepwalk” into a world where we delegate too much of our investigative intuition, or too much of our creative capabilities, or just focus on creating more and more content with less and less substance. It’s not quite Terminator and Skynet, but I think the detrimental impacts of sleepwalking are still pretty scary.
And now channeling your inner optimist, what’s one thing you hope to see for/with/from AI in 2025?
As a counterpoint to my sleepwalking fear, I have been buoyed by the ‘green shoots’ of more people seemingly being open to asking bigger questions triggered by widespread democratisation of access and the subsequent exploration of generative AI in particular. So, I am hopeful that as we continue to ask these big questions about who we are, why we do what we do, and how we best work with each other, that AI could really be used for good as part of the answer to those questions. Reducing toil, providing insights, enabling more accessible and intuitive interfaces - these can all be good things!
So, I’ll take that as my one thing…even though my Christmas 2025 wishlist for AI also has other things like helping with more effective and readily available treatments for insidious diseases, or significant progress in tackling climate and foodtech challenges. But you said just one thing. 😉
You can follow James on LinkedIn and subscribe to his Substack, Bridging Between Worlds. He is also an active speaker at future focussed technology events so keep your eyes out.