AI: Email, Flower or Frog?

Experts differ about its impact but the biggest companies in the world today are all investing billions into AI.

One of the first times I really noticed Artificial Intelligence (AI) in action was in the Google Inbox app, the Mountain View giant’s technologically-advanced version of Gmail. It suggested ready-made answers to an email I had just received. They were very relevant answers — one of them being exactly what I would have typed. I felt a tad guilty selecting the canned reply, thinking it was a little impersonal and a lot convenient.

While you can see more AI around, in eerily accurate Amazon and Netflix “you might also like this” recommendations; widespread coverage of DeepMind learning to play Go better than any human; and in self-driving cars that roam the streets of Silicon Valley, the next time I was personally aware of the progress of these technologies was when I took a photograph of a flower and Google Lens informed me it was a “Bird of Paradise”. Suddenly, technology was able to “understand” an image and determine its contents, which is exponentially harder than understanding text in an email.

AI, after many false dawns, is becoming very real. A concerted effort from academics and corporates, the availability of massive quantities of data inputs for machine learning and rapid advances in processing power have brought us to an inflection point. AI and its various subset technologies are now sufficiently advanced to take on high-level decision-making tasks previously the preserve of humans — and to do them faster and more accurately than we can.

Yet, outside of tech circles and some dramatic dystopian headlines, I don’t so far see much awareness of just how big a shift AI will cause. This strikes me as akin to the parable describing a frog being slowly boiled alive. The premise is that if a frog is put suddenly into boiling water, it will jump out, but if the frog is put in tepid water which is then brought to a boil slowly, it will not perceive the danger and will be cooked to death. After years of slow burn, where AI failed to deliver on its promise/threat of the science-fiction-style overthrow of humanity, the metaphorical water is getting very hot indeed.

There are those (such as Elon Musk and Stephen Hawking) who argue that we need to hop out of the water before it’s too late, or at the very least take steps to control the rising temperature. Others, such as Facebook founder Mark Zuckerberg, believe that the benefits outweigh the risks and we can reap massive advantages in areas such as medicine. Away from the philosophical debate among such luminaries, there are undoubtedly already strong commercial use cases and progressive businesses will rapidly gain an advantage by identifying how AI relates to their activities.

AI and the debates about it may seem slightly abstract for those embroiled in the pressures of everyday operations. But make no mistake, huge advances in the technology mean it’s already in our pockets, even if we don’t realise it. AI tends to be used as a buzzword or blanket term. It’s important to take the time to understand (at least at some level) the subtleties of the constituent techniques — e.g. machine learning vs neural nets. There’s no need for most people to become an expert but blithely referring to “AI” is analogous to a CEO saying she knows “business” without being able to differentiate between finance, marketing and logistics.

Should we hop away from AI? I personally don’t think so, but I do believe we need to give some serious thought to how we want to harness this next generation in computing capabilities. By not asking questions about artificial intelligence and its related fields, we relinquish a massive amount of control of the future — risking missed opportunities and preventable disasters.

— — —

If you have half an hour and want a good intro to AI, check out: https://a16z.com/2016/06/10/ai-deep-learning-machines/