What if it all goes 'Pop'?

There's a lot of talk about an 'AI Bubble' and the potential for it to burst. The market feels HOT and there's some things in there that don't necessarily stack up. But when the Bank of England starts predicting it, then it's probably time to at least start considering the consequences.

There's plenty written about what this means if you're invested in tech stock. But what about the many who aren't, but are concerned about what this means for the progress of AI technologies? I've been doing some reading and podcast listening (links at the bottom!) and have consolidated my thoughts into a few sentences which I think I have at least some degree of confidence in.

That said, I most certainly don't have a crystal ball, nor am I an economist or investment analyst, so please challenge my assumptions, tell me where I might be wrong, or what I might have missed... and definitely don't make any investment decisions from it!

So what if it did burst?

1. Historic 'bubbles' (from Rail in the 1840s to dot.com in the 1990's) suggest that what survives is good infrastructure and the strongest use cases. We very much continued using railways and the internet, both have had immense value and driven innovation long beyond their respective 'pop'.

2. AI technologies may face a similar shake-out, but much of the infrastructure, and all of the underlying technology and recent capability advances will still exist.

3. Most organisations haven’t scratched the surface of what’s possible with current AI technologies. Even if advances slowed (which is not necessarily an outcome from a bursting bubble), there is a significant amount of value to be obtained from well thought-through application of existing AI technologies.

4. The 'impact radius' of the economic fallout may be greater than dot.com (where the greatest impact was felt by those directly invested in the stock, but with some notable second-order impacts too) but probably not as great as the housing-crisis which caused wide-ranging impact on the global economy and living standards.

5. Watch out for investments that you think might not be in tech, but are connected (in non-obvious ways) to some of the big AI infrastructure projects eg construction of data centres... so if you're managing investment portfolios (looking towards many of the funders/philanthropic organisations that we've worked with), now might be a time to review these for connections to AI CapEx projects.

What's wrong? What did I miss?

Further Reading

'The bubble may well burst' - Guardian

'The bubble may not burst' - Goldman Sachs

One of the clearest explanations of some of the market forces at play that you could wish for (podcast) - Plain English

Different perspectives on the bubble... from Prof Galloway @ Bubble.ai

(Note: AIConfident exists to support leaders of organisations who may not have previously had much to do with AI technologies. This post is targeted towards them. The points above would be quite different for organisations deeply invested in tech stock!)

--

Image credit: Hanna Barakat & Archival Images of AI + AIxDESIGN / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/