AI Mania Is Eviscerating Global Decision-Making

Do yo have links handy for AI-assisted engineering going positively? The case I have on my mind of it going negatively is this recent Ford case [1] It's not that I believe it couldn't go positively, of course.

[1] Ford rehires human engineers after AI fails to match quality checks

https://www.bbc.com/news/articles/cgrkd41n2v9o

In this moment, the opposite is happening. Everything is getting called "AI", whether it uses LLMs to prompt LLMs about how to prompt LLMs, uses "conventional" machine learning, or just looks mysterious enough that they can expect the market to not ask questions.

I am reminded of "game AI", which for the most part has historically been just giant decision trees, encoded one way or another, because if you hook up any sort of real AI to a game entity or collection of game entities that does any sort of learning or training, even simple 1980s-era reinforcement learning, it turns out the game entities will roflstomp the human players, and the human players aren't interested in paying for that experience. We've been calling those collections of if statements and for loops "AI" for a long time, though, because who wants to hear about how deliberately stupid their opponents are?

> All of the AI projects we have observed as a team are failing. Every single one – we have seen 0% success in a year and a half, not only amongst projects we have been asked to participate in, but even within projects that we have observed in passing while doing totally unrelated work.

That's got to be hyperbole, which blows out their credibility. They chose to say 'AI' rather than, for example, LLM, or Transformer model, or Diffusion model. This means they are including a huge swathe of things dating back to Expert Systems in their claim.

And who hasn't seen productivity gains from more established AI technology - at least things like semantic search? Who hasn't seen diffusion models generating content in roles that might have done the work by hand before? Who hasn't seen some kind of regression algorithm (even using linear regression in a supervised context counts as AI - so you can absolutely do AI even in tools like Excel) improve operation productivity?

Even if they narrowed it to the Transformer model LLMs which re-ignited recent public interest in AI, less ambitious projects to give them to engineering staff to automate easy but boring tasks in the background generally have been a success. More ambitious ones that are beyond what you'd reasonably expect the models to be able to do - for sure, those tend to fail. For most of these, the failure is predictable in advance, while some are at the boundary of what's possible, and so it is harder to predict (these are rationally genuine R&D projects).

You are basically calling out the fact that the Emperor has no clothes. Many said this before. While it is a true statement, it is not going to help. Because, as you rightly said, it is a mania - like the tulip mania of 17th century or the manias of many forms today. The mania continues to evolve and flourish through it's peak and then go down. For that matter, there is hardly anything that is not a mania. Think of agile processes, timesheets, LoC based productivity, ...

The corporate mindset keeps going through different mania at different times. It could be initiated by some consulting gurus (processes), or some security nerds (strap yourself down until you can't move), or peer pressure (fear of missing out), or presentation goals (show that you are a AI-powered and modern company).

We can't remove or stop manias. Infact that is not the goal. The music should go on and the dance should go on. Everyone is in this dance - customers, businesses, supply chains, governments, thinkers and philosophers. It's a world-wide dance. So it's OK. The music track won't last forever. It will change and dance will change.

Clicking the footnote for the weird "All of the AI projects we have observed as a team are failing" is equal parts enlightening and confusing:

> We have rejected all AI implementation work. It is absolutely a gigantic bubble and we have minimized our exposure to it – every single one of our current contracts would be totally unaffected by OpenAI collapsing, save for perhaps some second-order effects such a recession causing a client to become unable to pay us. And there’s nothing we can do to insulate ourselves from that anyway.

Following the link to their company page goes to Hermit Tech, where the primary advertisements for their services are about helping failing projects and troubled teams.

So this is just one huge selection bias example? Start a consulting company for recovering struggling projects, then make claims like "100% of the projects we've seen are struggling"?

There's so much more in this blog post that feels like they're working hard to ignore anything that disagrees with their bubble. Building an AI data pipeline with evals such that you can swap between AI APIs is standard. It's actually part of doing a decent job because you need to select which model hits the right cost/performance tradeoffs and be in a position to pivot when that math changes. Harboring ideas that OpenAI is going to collapse and bring your projects down with it is the kind of talk you hear out of people who don't understand how AI projects work or that there's an ecosystem to it beyond a single company.

The latest projects I'm working on even include open weight models that can be run on reasonable local hardware as cost and performance benchmarks. Even if all of the AI providers collapsed at the same time and nobody offered any services (not going to happen) these projects can still continue on.

It's a very weird time in technology. You can have one foot in a world where people are adopting technologies and using them intelligently, then you can run into articles like this from people who have built their own little self-selecting bubble that confirms all of their ideas who can't even imagine that successful projects exist right now.