Kevin Scott: AI needs to be as accessible as the internet to reach its full potential

Kevin Scott: I grew up in a small town in the United States – Gladys, Virginia. When I was in my teens, I got super interested in computing. I was fortunate enough to be able to explore that passion, which translated to eventually earning a degree in computer science. After almost turning to academia and becoming a professor, I decided to take a job at pre-IPO Google to help build out their engineering office in New York.

From there, I joined a startup called AdMob and then landed at LinkedIn before their IPO to help scale their technology team and operations. Three years ago, I joined Microsoft; today, my job as CTO is helping the company set the technical vision to achieve Microsoft’s mission to empower every person and organisation on the planet to achieve more.

There are so many exciting things happening in technology right now. We have established technologies, like the cloud and mobile, becoming cheaper, more powerful, and more accessible to a broader range of people.

But perhaps the most important are emerging technologies like AI, synthetic biology, and edge computing where breakthroughs in recent years have expanded the range of problems that humans can solve, and that are advancing at such an incredible pace that they will undoubtedly help us tackle dauntingly hard problems we face in the future, such as climate change, feeding a growing population, dealing with future pandemics, making high-quality healthcare universally accessible, just to name a few.

I think that the emerging technology that will most move our world forward in the coming years is AI—as much if not more than electricity did in the 20th century, and the steam engine did in the 19th.

In order for AI to reach its potential, for it to become a tool that we all can use to help solve problems and create opportunities in our own communities and businesses, it will need to be an inclusive, accessible platform, much like the internet, that we can build anything we can imagine on top of.

And this platform will need to be powered and delivered at a truly unprecedented scale and democratised so everyone can innovate and benefit.

It’s hard to do. Part of the reason that many of us work in technology, and why technology products and trends can have a fandom around them, is that they can provide a compelling vision of the future.

Sometimes the more compelling the vision, the easier it is to suspend disbelief, and to get caught up in hype that has little substance and that is unlikely to ever create any disruptive impact. Striking the right balance between a belief in a future you would like to see realised, and scepticism about the path you are following to get there is really important, especially for the folks building disruptive technology.

Too little faith in the destination and the path you are following to get there, and you will fail to get anywhere at all. An inability to recognise when the path you are following is leading you to the wrong destination may mean that you never reach your goal, or worse, that you reach your goal and you realise that you don’t want to be there.

The scientific method isn’t a bad way to strike this balance: clearly stating hypotheses; carefully designing and running experiments to test those hypotheses; exposing your results to rigorous scrutiny by experts whose allegiance is to truth-seeking, and adapting yourself to what you learn from your own experiments and that expert scrutiny.

Practice compassion. The larger and more complex human society becomes, the more interdependent we become on one another, and the more we need to come together to do the important work ahead of us. It is hard to imagine how we depend on and trust one another enough to do that work if we don’t start by trying to truly see one another.

When I was a child, I loved to visit my grandfather’s repair shop, where I developed a deep appreciation for building and fixing things, problem-solving, and the scientific method. From there, it wasn’t much of a leap to teaching myself to program as soon as I could get my hands on a computer—a Radio Shack Color Computer II.

I’ve been endlessly fascinated by the possibilities of technology as a gateway to exploring curiosity and solving difficult problems ever since.

I get up at 5 am, exercise, do e-mail and catch up on what happened while I was asleep, eat breakfast, and after our kids are launched to school, I am in the office by 8am. Even before the Covid-19 shutdowns, I worked remotely.

I live in California for the moment and Microsoft’s headquarters are in Redmond, Washington. I spend a bunch of time meeting with people, individually or in groups, via video in Microsoft Teams. Most of those meetings are about technical work that we’re doing. I try to set aside a significant amount of time every week for deep work: reading scientific papers or technical reports; writing; coding; and experimenting. I head home by 7 pm, eat dinner with my wife and kids, and maybe watch a show together. Last night it was an episode of The Great British Bake Off, which we all enjoy.

The kids then head to bed, and I do one last pass at e-mail. I then try to read or watch something that teaches me something new before I head to bed. Last night I was watching videos about circuit design for constant current switching regulators for an LED lighting project I’m working on in my spare time.

I like to make things and to learn how things are made.

John von Neumann. He had a staggering breadth of curiosity and accomplishment.

I think that humanity is facing a bunch of problems where the solutions will depend on technology. I already mentioned a few – climate change, healthcare, expanding the food supply.

The technologies we will want to use to solve these problems, particularly AI, synthetic biology, and edge computing, need to be made as widespread and as accessible as possible so that we can have the full measure of human ingenuity employing them as we face our future.

In a sense, our biggest challenge is the will to invest in these things, and to educate and train as many people as possible to both participate in the development of these technologies, and to use them in service of a more prosperous future for us all.