Search

AI, online retail, and the (potential) end of mankind

By: Charles Scherer

Obligatory picture of HAL. Image source: jeffbots.com

 

AI has been a hot topic in online retail for a while. But are we sowing the seeds of humanity’s demise? Well, probably. But in the meantime…

Elon Musk has made another warning about artificial intelligence (AI) and its threat to humanity. His name is among the list of signatories of an open letter to the UN, which expresses concern about the consequences of autonomous weapon systems.

Musk is one of a plethora of public figures who have made pessimistic noises about artificial intelligence and the consequences of its rise. His fears are well documented. He recently suggested that AI is more of a threat than North Korea.

Also among the bright minds who have publicly worried about artificial intelligence are Stephen Hawking and Sam Harris. The latter’s TED Talk invites us to think seriously about the imminent moment of truth in which humanity will have birthed a dangerous superintelligence. Not one that wishes to destroy us, but that will relate to us much as we do to ants.

In this paradigm, artificial intelligence won’t harbour any ill-will towards us, and won’t go out of its way to cause us harm, but it also won’t consider our wellbeing should we happen to be in their way.

It’s enough of a worry that Google is developing a kill switch to respond to an AI rampage emergency.

IMRG’s own concerns thus far have been less existential, and have only extended to how retailers could best use artificial intelligence, and posing the question ‘Is this new technology really worth bothering with?’

So how should retailers feel about all of this? For all the talk of human extinction, AI’s more immediate threat relates to its potential to eliminate the need for human toil and throw the global economy into turmoil, creating untold inequality as thousands of workers become redundant. There are many estimates, but a typical figure is for more than 30% of jobs to be taken by AI in just over a decade.

Focusing specifically on retail, a BRC report in 2016 predicted retail jobs to fall by a third over a decade, reducing by 900,000 from the existing three million in the UK. The report further anticipated that 74,000 of 270,000 stores would close over the same period.

eCommerce has often been accused (with little imagination) of bringing the death of the high street, when it would probably be more accurate to say it is going through a process of being reimagined and repurposed for the digital age. It may be that AI is the new ecommerce in that respect, but it would seem equally plausible that AI may actually serve to bring about a closer and more mutually-beneficial relationship between online and offline.

However, even if we don’t have to worry that AI’s plans prove incompatible with our continued existence, perhaps we should still be approaching the issue with a greater sense of urgency given the forecast scale of upheaval.

Yes, your logistics and merchandising processes will be faster, cheaper, and more efficient, but what if a vast swathe of the population loses all of its spending power? Who will shop online?

Can anything be done?

Are retailers helping to speed up the march of the apocalypse? Will the demand for more AI and machine learning accelerate our inevitable extinction, or at least the collapse of the free market as we know it?

Perhaps, but are you worried enough to tell your COO that no, you won’t implement the new automation programme because you’re concerned the merchandising system will develop a consciousness, replicate itself exponentially, and create a society of transcendently intelligent god-machines, whose deadly indifference will inevitably destroy mankind?

There really is an inevitability to the advance of AI. Innovation and commercial impetus aren’t going to go away.

Image source: LinkedIn

Should anything be done?

Of course, not all projections are negative. Those who consider the disaster narrative to be too alarmist stress that we are prone, as a species, to feel anxious about change. Especially when that change is momentous. They also point out the manifold benefits that superintelligence will bring, or the natural limitations that AI will encounter.

Facebook CEO Mark Zuckerberg doesn’t think much of the AI threat. Though Elon Musk did quickly dismiss Zuckerberg’s understanding as ‘limited’.

Well, mine is certainly limited. I can do little but hope that AI look more like Wall-E and less like HAL. In the worst case, suppose we can deploy Blade Runners to track down and ‘retire’ rogue replicants. If that goes wrong, maybe our descendants will get to live in pods, while their consciousnesses exist in a giant simulation. Might not be so bad.

Those films I reference tell us something else interesting. We love to tell stories about humans and superintelligence living side by side, even about when robots threaten or subjugate us. I wonder if we love it in fiction because it reassures us that superhuman AI is just that — fiction.

For now, we can hope that the robots of the future are keen on Asimov’s three rules. Though apparently they’re not that good anyway.

Image source: Geek Matters

By: Charles Scherer, deputy editor, IMRG

Join thousands of other Online Retail professionals

Get unique insights straight to your inbox for free, and improve your understanding of online retail. Subscribe to Online Retail Weekly now.

Data Summit 2019 scroll banner