Get in touch

Do you know your AI from your elbow?

I read a very interesting article in the Financial Times the other day. “Retraining workers for the AI world” is the title and it began by quoting a recent announcement from the chief executive of JPMorgan Chase’s Asset and Wealth Management business: “This year, everyone coming in here will have prompt engineering training to get them ready for the AI of the future” hailed Mary Erdoes.

Now I confess that I have been struggling for a while to get my head around the idea that “prompt engineering” is a thing, but it seems I may not need to struggle for much longer. At the end of that very same article was this claim from Erik Brynjolfsson, Stanford University professor and AI specialist : “Prompt-engineering skills were hailed as the important new skill to learn … But they are already being eclipsed, as [large language models] learn to write better prompts than humans.”

I appreciate that when it comes to AI the pace of change is almost unprecedented, but this article in the FT made me wonder if we are all “enthusiastic laggards” now. The adoption curve as we know it is dead.

The near constant stream of announcements and breakthroughs from AI vendors reflects the energy and investment being put behind what, for many, is a make-or-break development in technology. I haven’t seen a vendor frenzy like this since the dotcom bubble.

And it is generating a lot of excitement in public and private sector organisations up and down the land. I had a hypothesis. It went something along the lines of “All of this upheaval and uncertainty … I bet that businesses would rather AI had never happened”. But my hypothesis is wrong. Vanson Bourne’s AI Barometer for June shows that the vast majority of IT and business decision-makers in the UK are positive about what AI can do for their organisations.

That being said, the uncertainty around AI was evident in a large number of the comments received from those who completed the survey on which the Barometer is based.

So what do organisations do when presented with technology that has the potential to be transformative but where the applications are not well understood and the FOMO is tangible …? Enter the consultants.

How’s that strategy coming along?

Remember how the growth of cloud computing was anticipated to simplify operations and reduce costs for businesses by offering scalable and efficient computing resources? Remember how the complexities and challenges associated with migrating to and managing cloud environments effectively led to an absolute fortune being spent on cloud consulting services?

Vanson Bourne’s AI Barometer for May reported the proportion of respondents whose organisations are, to any degree, retaining external consultants to assist with developing their AI strategy. A whopping 69%. Never mind Nvidia’s stock price, AI consulting is where the real money is at.

My notes say <insert presidential quote here for impact and gravitas>. Here’s one from George W Bush: “There’s an old saying in Tennessee — I know it’s in Texas, probably in Tennessee — that says, fool me once, shame on — shame on you …” … err … never mind. You get the gist.

All of this expertise being deployed at no small cost can only lead to great things, right? Right? Not so much. A measly 10% of respondents say that they are “Very confident” in their organisation’s AI strategy. Something doesn’t add up.

To be fair, around 70% of respondents are at least “somewhat confident” in their strategy. But I’m “somewhat confident” that Spurs will win the Premier League next year so I wouldn’t put too much store by that. When I’m spending consultant money I want to be more than “somewhat confident” in the results.

We don’t know what we don’t know

It’s the uncertainty. People don’t know exactly how AI can help them. They don’t know what operational impact it will have. They don’t know how to roll it out. They don’t know how AI will impact on their jobs or the jobs of their colleagues. They don’t know what their company’s competitors are up to (and they fear the worst). And they don’t know how the technology is likely to evolve, which solutions to invest in for the long term or which vendors to trust (you’re not helping, OpenAI).

Are vendors reassuring them? Not really. Most of the announcements from AI vendors seem to be aimed at the stock market and not their customers. New capabilities, new ventures, new technologies all announced to great fanfare … and all adding to the uncertainty.

So where did we land?

At the time of peak dotcom bubble, I was working for one of the world’s leading market research firms. Highly profitable but with mediocre growth, it was what the people at Boston Consulting Group would call a “cash cow”. It wasn’t sexy. Dotcom was sexy though and so the firm I worked for embarked on a merger with a dotcom start-up that was less than 5% of its size and loss making. A merger! Fully one third of the seats on the new board were allocated to people from the start-up’s leadership team. It seemed crazy to me at the time. And it was crazy. The deal was unpacked within 3 months.

I suspect that not a lot of thought was given to customers and what they might make of it all. Has anything changed?

 

What questions are you currently pondering? If you’d like to know more about how our bespoke research services – tailored to the tech sector – can help, get in touch for a conversation.

To keep reading, discover the findings from our ongoing AI Barometer series and more on our insights page and explore the work we do with our clients on our case studies page.