Die behauptete Steigerung der KI-bedingten Entlassungen um 1.100 % im Jahr 2025 könnte irreführend sein – Unternehmen wird vorgeworfen, die KI-Leistung zu übertreiben, um schlechte Geschäftsergebnisse herunterzuspielen
https://www.tomshardware.com/tech-industry/claimed-1-100-percent-increase-in-ai-driven-layoffs-in-2025-might-be-misleading-firms-accused-of-exaggerating-ai-performance-to-downplay-poor-business-performance
10 Kommentare
Anyone who didn’t have a „No shit Sherlock“ type reaction to the headline needs to pay more attention because those public lies are obvious.
We’re quickly transitioning from a world in which things that were true used to matter more than things that were believed. Today it’s going to be challenging to validate most kinds of statements. The ideas of facts and beliefs are bleeding together in front of our eyes. It used to concern me. Not as much anymore. Let’s face it, we’re transitioning into a digital reality that at least in theory is whatever you want it to be.
Had someone told me that critical assessment would be this generation’s cursive. I would have just laughed. But the more I consider this the more I come to understand it just doesn’t really matter what is true or false anymore. Not when it’s all ones and zeros.
> might
Are you high? What do you mean, _might_?
This is probably true in some cases, but don’t get rug pulled when this really starts happening this year
The AI performance may not be there, but that doesn’t mean the layoffs won’t come anyway.
It is true that many layoffs are not due to AI replacing jobs. But the poor performance is often due to loss of customers to AI. Either way it is still AI driving layoffs.
I wonder if the jacking up prices is finally affecting consumer spending. Prices have been going up drastically since COVID.
I have noticed a sharp uptick in apps just not functioning correctly.
The brilliant app for example, doesn’t let you scroll all the way down to see all response options so it’s hidden behind an overlay that is automatically there with no clear way to dismiss it.
Google Maps does not remember your default settings for which music player your prefer. And reddit is often a mess.
I suspect this is largely due to all of these companies rushing their employees to ship features, so they resort to AI, and without proper QA/testing (which is the first team to go) you get a bunch of software shipped that’s half baked.
The craziest part in all of this is the principles haven’t really changed for app-building since the 2010s even if the frameworks have been updated, but they’re just so much more bug-riddled nowadays.
I mean we’ve seen CEO blame everything but themselves for earnings.
Current LLMs are not AI. No one knows currently how to build “AI”.