Is the goal of democratising AI just a pipe dream?

29 Sep 2016

If the social media revolution has taught us anything, it is that tech giants will need real humans to give machines their vaunted artificial intelligence. Image: Michal Bednarek/Shutterstock

Microsoft, Google, Facebook, Amazon, IBM and other tech companies have joined forces to democratise artificial intelligence (AI) and make it available for everyone. But what are their real motives?

The AI world is exploding. 10 of the biggest power brokers have joined forces to bring some clarity to a world still fumbling with smartphones, and one that hasn’t quite gotten its head around what AI really means for it.

It’s called the Partnership on AI to benefit People and Society.

Sounds lofty and earnest, but these are businesses after all. Could this be more of the passive-aggressive PR shtick from a bunch of tech giants who talk about being open and transparent, but really prefer to keep you at arm’s length behind security desks and email addresses that no one replies to?

It’s like talking up social media when your approach can at best be described as sociopathic.

What snake oil is this? Do they want to improve AI as a team when they compete ruthlessly against each other? What are they really selling here?

Remember dear human, you will always be the product

The tech giants pushing AI are already driving significant advances; one might say that we are in the age of machine intelligence.

For example, Intel recently bought Irish tech company Movidius, whose machine learning chip can help devices like DRI drones to see and understand.

It’s like we are at the gates of Florence and the Renaissance is about to unfold.

And it already looks magnificent.

Microsoft is building what it claims to be the world’s most powerful AI supercomputer, which it plans to make available to everyone to harness the full power of AI.

IBM’s Watson has been beating humans at chess for some time now.

Facebook has already installed AI agents inside Messenger to converse with humans on behalf of businesses.

Google’s latest neural machine technology can translate entire sentences from English to Chinese and vice versa.

As for Amazon, God only knows what Jeff Bezos has up his sleeve, but it probably involves swarms of delivery drones or entire AI servers on demand (think Amazon Supercomputer Services) that will unleash another wave of entrepreneurship.

But when big businesses talk about sharing the spoils of their advances in AI do they really want to share? What’s their real game here?

In the dawn of social media, it was information. What you do with that information, that is called intelligence.

It therefore makes sense to the masters of AI to invite society to their party good and early.

Think about it. Humans are already product, when you consider the 1.5bn people who use Facebook every month. Humans are advertising fodder. It makes sense for Google to create free services like Translate or Gmail because it’s a captive audience for advertising. It’s not personal – well, it is – but it’s just business.

Microsoft seems to be really taking this benefiting humanity stuff seriously. Not only did it promise free supercomputing resources, but it made a bold statement this week that it intends to infuse every application and device with cognitive capabilities. It has already done so in Windows 10 with Cortana, which it says has achieved 12bn queries from over 133m active users.

Other examples include the photo editing tool Pix, which has apparent powers of perception and a location-aware app called MileIQ that helps you calculate expenses without doing much.

Another thing worth considering is how AI could be used to replace humans in jobs. Who needs call centre agents when Facebook Messenger has bots that can carry a good conversation? Will AI reduce the number of people in the working world? Will robots take our jobs? What is the point in talking about democratisation when people won’t be able to put food on their plates?

Or will AI go the other way? Will it augment our lives, make us more productive and lead to more opportunities for gainful employment and fulfilling healthy, happy lives? Let’s hope so.

“AI technologies hold tremendous potential to improve many aspects of life; ranging from healthcare, education, and manufacturing, to home automation and transportation,” the Partnership on AI said in its mission statement.

“Through rigorous research, the development of best practices, and an open and transparent dialogue, the founding members of the Partnership on AI hope to maximise this potential and ensure it benefits as many people as possible.”

That’s fair enough. But you can’t help but wonder how AI is going to benefit these behemoths first before the rest of us figure it out.

John Kennedy is a journalist who served as editor of Silicon Republic for 17 years