Augmented Humans, AI and the State of the Art
I feel a bit sad for the State of the Art. Not for the current technological environment in which we find ourselves, but for the phrase itself. Once a cutting-edge[1] term in advertising and journalism, it quickly became overused, then a cliché, to now being a much-maligned and easily ignored bit of salesy jargon: state of the art fidget spinners! State of the art rompers for him!
We can trace the origin of the phase to the height of the second industrial revolution – the one in which technological advancements exponentially increased productivity and economic growth, and mass production brought new technology (such as the automobile) into the hands of everyday people. Thinking of the state of the art in this context, then, gives new relevance to the term as we find ourselves in the midst of a fourth industrial revolution – one where data is the new fuel for innovation. In the run-up to London Tech Week (June 12-16), we’ll be taking a look at the impact of state of the art technology on the way we’ll learn, live and work. In this series, we’ll also examine the art of the possible in the very near future.
State of the art-ificial intelligence: it’s about data
No technology represents the state of the art of the fourth industrial revolution better than artificial intelligence. There’s a tremendous amount of buzz about AI: people worried that AI will take away jobs or that it’ll be the magic bullet to solve every business and societal problem. At its most basic level, AI really boils down to data, and a practical set of tools, techniques and technology used to make sense of it. It’s about having the computing power available anywhere to develop an increasingly complex set of algorithms that allow AI to sense, comprehend and act within its environment. I’ll examine how this is applied in various industries in a later post, but for now, here’s an example of how the evolution of data is connected to the evolution of AI
Our cars have, for a few decades now, on-board computers that could record a fault and share that data with a mechanic. That’s a reactive use of data. Now you can add a small data collection device to your car so that your insurance company can predict how safe a driver you will be. That’s predictive. Moving to AI in its current state, Uber drivers are using facial recognition (a type of AI) to log into their systems, and a number of companies are using data to develop self-driving cars.
You’re already using it
In the context above, the state of the art for AI is much more practical than scary. In fact, you’ve probably already used AI several times today: from the machine learning capabilities in the spelling and grammar check of a Word doc you’ve written, to the facial recognition you may have used to log into Windows 10, to the natural language processing you used when you asked your phone to search for something. Behind the scenes, AI is hard at work doing things like keeping your devices safe as part of Windows 10 Advanced Threat Protection. In other words, AI is already part of the apps and devices you use every day.
Humans vs. and Machines: Augmented Humans
As AI continues to evolve, it will become more ingrained into the fabric of our everyday lives. While there’s a great deal of concern about the way AI will replace humans in the economy, we see the relationship with this technology as enhancing what we do, not replacing it. It’s very much machines and humans, not machines versus humans. A recent Accenture study on AI in 12 developed economies reveals that AI could double annual economic growth rates in 2035 by changing the nature of work and creating a new relationship between man and machine. Accenture estimates that AI could add over $800 Million (USD) to the UK economy by 2035. This benefit will come from UK start-ups such as Hippo Data, who use machine learning to provide an algorithmic trading platform, or Metail that provides a computer vision platform for retail / fashion industry.
Democratising of AI
As AI evolves, however, it’s important to make sure that it remains available to everyone, not just a select few. In the same way Microsoft’s vision to have a PC on every desk in the 1980’s was key to the broad adoption, we see the same need for AI to be democratised. It should (as with everything Microsoft strives to do) enable every person and every organisation on the planet to achieve more. That’s why we make much of our AI functionality available free for developers through our cognitive services APIs. As more powerful algorithms for AI make their way to our mobile and IoT devices (the intelligent edge), it’s important that the raw compute power existing because of the cloud is available to all developers. And while this broad availability of AI (from smart assistants to AI-enabling cloud infrastructure) enables developers to create the businesses of the future, this blog is primarily a look at where we are today. We’ll delve more into the art of the possible – where technology is taking us in the near future – in subsequent posts and at London Tech Week.
If you have a chance to attend London Tech Week – a festival of live events across the city, showcasing and celebrating the best of tech whilst providing networking, social, learning and business opportunities – please come visit our interactive Studio Spaces @ 110 Pennington Street, London or hear from Microsoft UK CEO, Cindy Rose, at the LeadersInTech summit.
Find out more about London Tech Week
[1] Another bit of overused jargon that’s in need of a comeback
