Baldur Bjarnason

Generative AI is the tech industry's Hail Mary pass

– Baldur Bjarnason
This page was originally published elsewhere, but is republished here for archival purposes.

I’m not one for using sports metaphors in my writing because my disinterest means I lack the context necessary for a nuanced use of a metaphor.

But, in this case, I think it’s the aptest one I could find.

A Hail Mary pass is a very long forward pass in American football, typically made in desperation, with an exceptionally small chance of achieving a completion. Due to the difficulty of a completion with this pass, it makes reference to the Catholic “Hail Mary” prayer for strength and help. (Wikipedia)

This may seem counter-intuitive if most of what you read is saturated with the tech industry pop culture, but that’s because the tech industry isn’t the most self-aware.

Over the past few decades, tech companies have been priced based on their unprecedented massive year-on-year growth that has kept relatively steady through crises and bubble pops. As the thinking goes, if you have two companies—one tech, one not—with the same earnings, the tech company should have a higher value because its earnings are likely to grow faster than the not-tech company. In a regular year, the growth has been much faster.

This has been great for the tech industry and for investors. Even shitty, poorly-run tech companies benefit because people assume tech will grow enough to make up for poor management. If you can demonstrate growth, investors will ignore that you’re pissing it away with reckless abandon. Tech companies—generally—have a baked-in assumption of future profitability that most other sectors don’t have.

The problem is that things that can’t go on forever don’t, and over the past year some investors have changed their tone on the tech industry. Gone is the assumption that endless growth will make up for the inefficiencies and incompetence. Instead, you have investors calling for massive layoffs.

Mass layoffs rarely make sense, and they almost never make sense for tech companies, but it is the default response of the US executive class to lean years, even though they have a tendency to exacerbate a crisis. This is fine for most executives because it’s a move designed to recover investor faith.

Temporarily, at least.

The core problem the industry faces isn’t waste but the possibility that their growth and economic value will revert to the mean: that tech will be seen as a stable, mature industry whose days of endless growth are behind it.

Tech executives have known this was coming for a long while now. It’s the reason for a lot of the so-called waste. These costly experiments were supposed to be it—the innovation that would keep the industry growing for another decade.

None of these panned out. None of these became the big bright future that tech needs to justify expectations of endless growth and the valuations that come with it. A few of these turned out to be partially or even blatantly illegal.

As Bruce Sterling put it in his 2023 State of the World:

Big Tech, in financial and technical decline across the board. “Detroit syndrome.” It’s hard to arouse raw enthusiasm for anything “digital,” because that high-concept is passe, it’s become like “electric,” “atomic,” “automatic” or “space age.” I’m a very “cyber” guy personally and I truly enjoy the industry’s antics, but I always knew there would be a day when “cyber” became cranky and old-fashioned tech – especially when adults came on to the scene who had never known anything else.

This brings us to generative AI.

Machine learning is an amazing field that has unlocked many of the most interesting advances in computing over the past decade. Apps like Descript make the impossible possible. The image recognition and extraction capabilities built into Apple’s OSes are fantastic. It gives Photoshop and other image manipulation software features that are incredibly useful. It lets special effects studios make magic.

That isn’t enough. Those are incremental improvements in productivity. Sustaining innovations, if you’re a Clayton Christensen fan, that increase the quality of a product but don’t introduce disruptive changes to the market.

If tech is to turn around its trajectory it needs something that will promise nothing less than the complete transformation of society—something at least as big as the internet.

And Large-Language-Models like ChatGPT are it.

Or, they’re supposed to be it.

As Bruce Sterling also said in his State of the World post:

Generative AI is quite interesting technically, it’s probably less of a big-deal than its zealots imagine. Also, “digital” has been over-extended into areas that are probably counter-productive – for instance, if Facebook ceased to exist, no one would mourn it. Alexa leaves the home, nobody notices. These examples could be multiplied.

What is important for you, and anybody who works in tech, to know, is that this move is desperate, even if the tech ends up doing what it promises. Tech is all in on generative AI because nothing else looks even remotely convincing.

This has a few consequences for the rest of us, the plebeian non-investors in tech, even if you do believe in the tech:

There will also be a push to use these tools to replace jobs. This isn’t the 90s anymore, where fast growth meant that automation and job growth could go hand in hand. Today, automation like ATMs, self-checkout, and online backing is not used to complement a company’s workforce. As soon as the 2008 economic crisis hit, automation became a tool to reduce an organisation’s headcount, not for increasing productivity or margins.

Knowing that the hype for generative AI will get desperate is helpful even if you do think it’s the greatest invention of the millennium because criticism and discourse is essential to improve it. People’s concerns and observations of the technology’s flaws are all opportunities to understand and improve it—make it work better with our society and economy.

If you believe in the technology, you should listen to its critics.

Instead, you are going to see emotional dismissals of any and all criticism.

Because AI is a Hail Mary pass that has to work.