So who has driven the 1000x increased usage of AI in the past year or two? My mother is in her 60s and uses Gemini every day. These data centers aren’t being built for no reason.
That's also because traditional google.com has become a product search engine instead of a knowledge search engine. So far at least, the AI results are mostly free of product placement and thus automatically 10x more useful than the first few pages of search engine results (but probably not for long).
I have many times now searched Google for an error message or similar and either gotten no results or been unable to get it to search for what I actually told it to search for instead of some vaguely similar but completely unhelpful phrase. The LLMs will find a link to a bug tracker or stack overflow. It’s crazy how much worse Google search results are now than they used to be.
Yes, Google search was good until they made it bad for profit and people still used it anyway because they'd lost the ability to do without it. We're in the era of being trained to rely on AI in the same way. If you think it will remain good, you haven't paid attention to the last ten years.
I don’t even think it’s good now. Sure, they’ll often give you a link. But sometimes they’ll just insist there definitely is a page that they’re pulling their made up info from but fail to provide a link, outright fabricate a link, or provide a reference that simply isn’t consistent with the purported summary. I’m sure the inevitable product placement won’t help matters.
> It’s crazy how much worse Google search results are now than they used to be.
I could not agree more. Each time I do a google search, I feel like I've stepped into a severe reality distortion field. The results are simply WORTHLESS. It's not just that the AI summary at the top answers the wrong question, but the sequence of hits thereafter wander off into the weeds almost immediately. I routinely have to constrain the search in multiple ways (time bounds, specific word rejection, etc) just to get one or two relevant hits.
Frankly, I think I'm going to have to MOVE AWAY FROM GOOGLE to a subscription-based search service just to get any useful search results again.
> Frankly, I think I'm going to have to MOVE AWAY FROM GOOGLE to a subscription-based search service just to get any useful search results again.
I would absolutely do this, but the only subscription search service i know about is Kagi, which is a bit too AI-forward for my taste. I wish there was a way i could subscribe to "google but 2007 google but with updated results"
If the AI results were so undeniably good, google would let me turn them off and let user preference prove it. I verify those AI results when i fail to avoid reading them, and they're wrong a shockingly high percent of the time.
> she isn't burning through nearly as many tokens as developers who are utilizing AI effectively
I don't think this takes codex into account, but the 'What they’re using it for' section shows there's definitely a huge demand outside the use case of programming:
https://openai.com/index/how-people-are-using-chatgpt/
At least partially the usage is driven by free plans. I use Gemini and ChatGPT for free. I will not pay for them unless traditional web search will be killed (google quality is subjectively on a downward trajectory for the last few years). My employer pays for AI but IMHO it's driven by a panic level FOMO, not evidence.