The Dismal Economics of Artificial Intelligence

Perhaps the single biggest reason Artificial Intelligence won’t reach the ominous scale of the Terminator or the romantic, dulcet tones of Samantha in Her or the childlike wonder of Spielbergs AI is economics. Although AI has made some truly astounding advances, I take a look at the more dismal side…because sometimes we have to.
Artificial Intelligence is a Business Proposition
While it may seem impressive that IBM, Google, Apple, Amazon and Facebook are dropping hundreds of millions into developing Artificial Intelligence systems, don’t ever forget that there is a business purpose behind these investments. They anticipate a profit. If that profit becomes elusive, AI research at this current scale will fizzle rather fast.
IBM with Watson; the two have almost become synonymous and I’d wager IBM may even change its name to Watson. If AI pays off (and it probably will.) When IBM changes it’s name to Watson, we can be fairly sure they see the economics to do so. IBM is perhaps the one tech company that has placed the biggest bet on Artificial Intelligence. I’ve made extensive use of Watson services in the past two years. It is impressive, as in very, very impressive. If I had to bet on a winner over Google, Facebook or Microsoft, I’d be firmly in IBM’s camp, followed by Microsoft.
The Cost of Thinking Machines
Getting to the Hollywood portrayal of an AI that is truly cognizant and aware is likely economically unfeasible. But then most companies investing in AI don’t actually want to get to the point of a machine that actually thinks like a human (one might argue that a machine could never think like a human simply because it is a machine.) It would take a significant amount of ongoing capital and resources to get there. Yes, progress is significant and there will be some excellent AI tools in time. Today, AI is equivalent to, at best, Windows ’98. Maybe only even MS-DOS.
The Business of Artificial Intelligence
On the one hand it’s easy to slap on the tinfoil hat and suspect governments of developing spooky AI systems for global domination. The reality however, is that is unlikely in the extreme. The economic value isn’t there, so governments will likely buy AI from corporations. So it all falls back on corporations. And economic viability.
The major companies investing in AI development are doing so to unlock the massive profit potential. And it could be massive. The smaller ones and startups working in the AI field are doing so probably with the intent to be acquired by a big player and that’s already happened a few times (i.e. Siri.)
AI – Bear or Bull?
Having helped a few companies determine the potential of leveraging AI for their business strategy, I’m personally bullish on AI. With qualifiers. Companies buying and using AI must have good ethical/governance behind it’s intended use. Secondly, integrating it is no small task and the actual value of productivity gains or cost reductions is not yet proven. Today, most AI tools are singularly focused.
What should also be considered is power systems in industry and government. At this time, there is a big enough challenge for CEO’s to defer decisions to Big Data analysis (many CEO’s are reluctant to give up on “gut decisions”.) Doctors are extremely reluctant to pass on decision making to automated tools as well. It is more likely in the next 15-20 years that as AI advances, it will find it’s place in low-level functions where the cost/benefit/risk scenario makes it more economically viable and poses less threat to senior management (such as highly process driven decision making where intuition is not much needed.) And it is highly likely AI won’t be feasible, for a very long time at least, to take on the human role of the C-suite and finer points of health care, international relations or global trade issues.
Ray Kurzweil, the renowned and brilliant futurist  with a long track record of being right suggests AI will reach the “singularity”, where man and machine become intellectually equal, within the next 15 or so years. I have no track record or brilliance like Kurzweil, but I doubt this will happen so soon. If ever. The economics seem to argue against it.
What do you think? Will AI be a profitable venture for business?
Footnote: I had the opportunity to hang out with Ray Kurzweil for a couple hours a while ago; I learnt quite a bit and gained some key insights. A brilliant man. Taken in Halifax.
Ray Kurzweil & Giles Crouch

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s