IT departments have a confidence issue with AI
Is anyone surprised by the level of concern expressed by Irish businesses about artificial intelligence in research from insurance broker and risk management company Gallagher in Ireland?
For those of you who may have missed it, Gallagher found 88% of Irish businesses were concerned about the increased threat of privacy violations and data breaches AI could bring. Exactly the same number worried about the potential for AI to produce misleading or incorrect information.
Every one of the top six concerns about AI identified by Irish businesses scored very highly (as in over 80%). They included algorithm bias and discrimination, liability or legal accountability in the misuse of AI, greater vulnerability to cyberattacks and fraud and a lack of skills within the organisation to leverage AI.
I have no argument to make about any of those anxieties at all. I think we’ve all seen quite enough to be cognisant of the fact that AI is vulnerable on all of those fronts. Or rather, the organisations deploying it are – as are the people interacting with those organisations.
Funnily enough, the one group of people who are always most sanguine about these particular concerns are those who are selling the technology. A number of them are currently accompanying President Trump’s Gulf deal tour, notably Sam Altman of OpenAI and Elon Musk along with representatives from Nvidia, Amazon, AMD and Alphabet.
In that vein, I suppose it should come as no surprise that, according to Laura Vickers, managing director of Commercial Lines in Gallagher, the survey found that “IT businesses were amongst the least concerned about AI” (although it should be noted that 50% still expressed concerns about it).
She argued their higher level of confidence could be attributable to their greater familiarity with AI which made them “better able to understand and manage AI”.
This is not quite as convincing as she might think given that Altman and Musk are two of the most confident men when it comes to AI but neither inspires confidence about it in everybody listening to them.
Foresight
But having made her first assumption, Vickers can’t really be faulted for taking it to the next logical step by arguing that if Irish businesses “learned more about this technology, it may help them to overcome their fear around it – and it may also empower them to make the most of the new technology, while also avoiding the risks and dangers it might bring”.
The assumption here, of course, is that the people promoting AI are doing so from a position of benevolent foresight and wisdom about its benefits and that this elevated perspective makes them far less anxious about its potential perils.
But here’s the thing. I’m not a salesperson but I know that one of the basic tenets to greater success in sales is to believe in what you’re selling. And that’s great and all but the problem with belief is it’s, well, it’s a view, an opinion, something that is accepted as true on the basis of inconclusive evidence.
And the difficulty with belief is that it doesn’t have to be tethered to reality. People can believe that AI is not as risky as 88% of Irish businesses think it is, but that doesn’t mean they’re right. It doesn’t mean the 88% are right, either, but they’re not the ones that need to convince anyone that AI does not have the potential to produce misleading or incorrect information or that it will include algorithm bias and discrimination.
So, I think what we have here is a very clear demonstration of what separates belief from proof. Belief is great if you’re the person selling something but proof is so much better if you’re the one being asked to buy it.
Subscribers 0
Fans 0
Followers 0
Followers