Welcome • User Guide
ToS • Privacy • Canary
Donate • Bugs • License

©2025 Poal.co

902

They’re starting to figure it out.

On the other hand, smaller, smarter companies might be figuring out legitimate uses for AI. Their adoption rates are already lower, so they probably didn’t take the “Force all of the plebes to use this” strategy.

[direct link](https://www.apolloacademy.com/ai-adoption-rate-trending-down-for-large-companies/) They’re starting to figure it out. On the other hand, smaller, smarter companies might be figuring out legitimate uses for AI. Their adoption rates are already lower, so they probably didn’t take the “Force all of the plebes to use this” strategy.

(post is archived)

You are currently inside a comment thread.
Click here to see all the comments (6).
[–] • 0 pt

It doesn't surprise me. They've run out of training materials, and are scraping the bottom of the barrel with preddit.

The next "big thing" will be to digitize everything written from the 20th century and earlier. At some point the bubble will pop or at least correct.

[–] • 0 pt

It won’t matter how much new training text they feed into them. They have trained them on terabytes of data. A few more terabytes is not going to fundamentally change an LLM into something that can think. They have reached the point of sharply diminishing returns. The latest ChatGPT models have even been worse than previous ones.

[–] • 0 pt

Grok is also worse than before. When they put kosher guardrails on, you notice the quality go to shit for everything. They have maybe 20% left to squeeze out of it.