The hed here is a bit misleading … 40% of total staff is not 40% of an individual job.
Jack Dorsey cited AI as the driving force behind cutting 40% of his company’s employees, but other factors such as a weak crypto market, overstaffing and a declining stock price may also have motivated the move.
Last week, the financial technology company Block announced that it would lay off 4,000 of its 10,000 workers. Dorsey, Block’s CEO, said in a letter to shareholders that advances in AI “have changed what it means to build and run a company”.
“We’re already seeing it internally. A significantly smaller team, using the tools we’re building, can do more and do it better. And intelligence tool capabilities are compounding faster every week,” he wrote. He also said that Block’s business remained strong and that these cuts weren’t an austerity measure.
Can AI operate 40% of a business? Perhaps, but other specters haunt Dorsey’s company.
I’d be surprised if LLMs can handle 40% of anyone’s job. You know what often can? Good, old-fashioned automation. It handles tedious tasks no one wanted to do in the first place and produces improved, predictable and testable results.
Remember boys and girls, AI doesn’t have to be as good as you to take your job. Just has to be good enough. Because AI doesn’t go home for the night, it doesn’t take vacation, it doesn’t take sick days, it doesn’t care about holidays, it doesn’t go to the bathroom, it won’t go get a cup of coffee, it doesn’t need pay and benefits, it doesn’t require payroll taxes be paid…
And now think about how the economy seems to have shifted to catering to the wealthy. Vegas only wants the high rollers, everyone else fuck off. Car companies have priced all their vehicles like luxury vehicles even if they didn’t do anything to make them luxury. Companies know only the wealthy will be able to buy anything in the near future.
The primary financial issue with LLMs taking people’s jobs is in the cost of operation, mainly for the LLM companies. They still have not made an ROI, not even close, and that’s with massive government contracts. I personally think that there’s one of three possibilities here. Most of the LLM companies could go under (but they may be “too big to fail” at this point), the LLM companies could start to charge far too much for the quality of the outputs causing some companies to back out, or the LLM companies will, by some miracle, get a ROI through a more efficient model.
The models are enshittified at the start, since they’re basically just hallucinating with guardrails. There’s not any way to make them truly deterministic. Even the “agents” that run through multiple iterations of code to find the “best” solution are lacking. This is because they cannot “think” logically.
My personal opinion, though, is that they’re simply using it as an excuse to fire workers and pump up company stock value. I don’t believe they actually think that LLMs can fully replace devs and engineers. So yes, while LLMs are taking jobs right now, it’s not because they’re good enough at what they do or anything like that. It’s just greed.
The price of these services is pretty heavily subsidized today. Unless their costs go down a lot over the next couple years , they will eventually have to raise prices which will destroy roi.
Agreed. I don’t see LLM services making an actual ROI any time soon unless something drastic changes.
It took longer than I expected, but Humans need not apply is becoming more and more relevant. Interesting that he thought it would be the transportation sector first that sees significant job losses. Instead that still hasn’t really taken off yet at scale and its the later predictions that we are now seeing.
What do we do when large sections of society are unemployable through no fault of their own?
To quote Oversimplified “Who wants to start a rebabablution, rebolalution… Dang it”