Meta is axing 600 roles across its AI division

I just assume they over hired. Too much hype for AI. Everyone wants to build the framework people use for AI nobody wants to build the actual tools that make AI useful.

Maybe because there are just very few really useful AI tools that can be made?

Few tools are ok with sometimes right, sometimes wrong output.

Integrating LLMs with the actual business is not a fun time. There are many cases where it simply doesn't make sense. It's hard to blame the average developer for not enduring the hard things when nobody involved seems truly concerned with the value proposition of any of this.

This issue can be extended to many areas in technology. There is a shocking lack of effective leadership when it comes to application of technology to the business. The latest wave of tech has made it easier than ever to trick non-technical leaders into believing that everything is going well. There are so many rugs you can hide things under these days.

There is a real question of if a more productive developer with AI is actually what the market wants right now. It may actually want something else entirely, and that is people that can innovate with AI. Just about everyone can be "better" with AI, so I'm not sure if this is actually an advantage (the baselines just got lifted for all).

Hmmm new business plan - RAAS - Rugs As A Service - provides credible cover for your departments existance.

There are N useful AI tools that can be made.

They’ve done this before with their metaverse stuff. You hire a bunch, don’t see progress, let go of people in projects you want to shut down and then hire people in projects you want to try out.

Why not just move people around you may ask?

Possibly: different skill requirements

More likely: people in charge change, and they usually want “their people” around

Most definitely: the people being let go were hired when stock price was lower, making their compensation much higher. Getting new people in at high stock price allows company to save money

And once the business inevitably files for bankruptcy it'll be the biggest rug pull in corporate history.

I don't know if this is true. It's good for some things... Learning something new or hashing out a quick algorithm or function.

But I've found it leads to lazy behaviour (by me admittedly) and buggier code than before.

Everytime I drop the AI and manually write my own code it is just better.

Where N is less than infinity.

Is it known that there are fewer than infinity tools?

For any given time period N, if it takes > 0 time or effort to make a tool, then there are provably less possible tools than infinity for sure.

If we consider time period of length infinity, then it is less clear (I don’t have room in the margins to write out my proof), but since near as we can tell we don’t have infinity time, does it matter?

> Integrating LLMs with the actual business is not a fun time. There are many cases where it simply doesn't make sense.

“You’ve got to start with the customer experience and work backwards to the technology. You can’t start with the technology and try to figure out where you’re going to try and sell it.” — Steve Jobs

I would assume that for any given tool you could make a "tool maker" tool.

> More likely: people in charge change, and they usually want “their people” around

Also, planning reorgs is a ton of work when you never bothered to learn what anyone does and have no real vision for what they should be doing.

If your paycheck goes up no matter what, why not just fire a bunch of them, shamelessly rehire the ones who turned out to be essential (luckily the job market isn't great), declare victory regardless of outcome, and you get to skip all that hard work?

Nevermind long term impacts, you'll probably be gone and a VP at goog or oracle by then!

I haven’t even thought of Meta as a competitor when it comes to AI. I’m a semi-pro user and all I think of when I think of AI is OpenAI, Claude, Gemini, and DeepSeek/Qwen, plus all the image/video models (Flux, Seedance, Veo, Sora)

Meta is not even in the picture

This is true, but sadly the customer isn’t always the user and thus nonsensical products (now powered by AI!) continue to sell instead of being displaced quickly by something better.

How convenient: the AI boss, LeCun just is not interested in that stuff!

VR + AI could actually be kinda fun (I’m sure folks are working on this stuff already!). Solve the problems of not enough VR content and VR content creation tools kind of sucking by having AI fill in the gaps.

But it is just a little toy, Facebook is looking for their next billion dollar idea; that’s not it.

You make a tool, then a tool factory, then a tool factory factory, ad infinitum.

My voice activated egg timer is amazing. There are millions of useful small tools that can be built to assist us in a day-to-day manner... I remain skeptical that anyone will come up with a miracle tool that can wholesale replace large sections of the labor market and I think that too much money is chasing after huge solutions where many small products will provide the majority of the gains we're going to get from this bubble.

Can you rehire that quickly though? I know where I live the government won't allow you to rehire people you just fired. Because the severance benefits have lower tax requirements and if you could do that you could do it every year as a form of tax evasion.

You should read https://www.fimfiction.net/story/62074/friendship-is-optimal.

Even tho the creator says LLMS aren't going in that direction it's a fun read, especially when you're talking about VR + AI.

Author's note from late 2023: https://www.fimfiction.net/blog/1026612/friendship-is-optima...

Are you in California?

There is no ASML toolmaker maker.

>My voice activated egg timer is amazing.

Alexa?

Not yet, but could there be?

Sprinkle in minimization and virtualization and it's extremely cool!

No this was in Europe. I would never work in the US, not even California.

VR + AI synergies is why meta released their model open source Im guessing. The other big tech companies largely have LLMs as substitutes to their products(google being worried about people using chatgpt instead of traditional search) but for meta their products have incredible synergy with AI.

Billion is too little for them tbh.