LLMs went from a toy to highly capable (along many dimensions) in just a few years.
Which makes it very likely that in a few years more, we will be seeing crazy capabilities that leave us humans increasingly behind.
Cynics really say stuff like "but it's been 3 years and all they can claim is that it's changed text to speech, speech to text, photos to text, text to photos, programming, cybersecurity, research, mathematics, war...
"They lie about it being dangerous!" and simulataneously "They hacked a system with zero day exploits! Everyone involved should goto jail!"
Tech leaders are awful, but it is just that we've built a society that enables the most awful among us to benefit the most. It has never been the case that tech leaders are particularly amazing, they are just amazing at exploitation.
FWIW:
The simultaneous "it's stupidly weak" and "it's dangerously strong" is a well known propaganda technique.
Instead of expanding the list, it would be better to widen your vision. Email was more transformative than AI is right now. Unix, then Linux was more transformative. If you have a magic wand and remove AI, I think barely anyone would be impacted, but removing something like windows would be like erasing computers. And removing Linux would wipe most of the internet services.
Unix in it's first five years reworked the way operating systems performed. I'd agree it's pretty important, but I don't see how it fundamentally changed the way we perform work in multiple disciplines (in its first 5 years). Same for Linux, but even less so.
We're in very early times, and generative AI is touching every corner of the workpace and education, for better and worse. Powerful technologies can have negative effects. Email has been negatively transformative in ways as well.
Part of the reason you can erase AI so effectively with so "little" effect is that it is new.
> We're in very early times, and generative AI is touching every corner of the workpace and education, for better and worse.
Forcefully. By a mixture of free offerings and bundling AI in every tools with no disable button. It's still not an essential tech like email or Linux is (not even essential at the level of GitHub).
> Forcefully. By a mixture of free offerings and bundling AI in every tools with no disable button. It's still not an essential tech like email or Linux is (not even essential at the level of GitHub).
We could survive without computers and the internet. We did it. I was there. We could go back. It's not essential. You've lived maybe 3 years with LLMs. No kidding they could disappear. As an old person, I have no idea why you think any other technology is any different. LLMs are new, that's the big difference as I said earlier.
> Forcefully.
So.... I hate mobile apps. I hate that TVs connect to the internet. I hate that we have cameras on every corner. I hate that we have cameras that watch us while driving.
Now, tell me about how this is different than all that.
As late as 2009, I was living without computers and internet. The most tech thingies in everyone home were the radio, the TV, the VCR (not everyone and half of them wasn't working properly) and a dumbphone.
I'm not saying that internet or any of the others is essential (in absolute). What I'm saying is that the supposedly "transformative" aspect of AI needs free offerings, dark patterns, and a mountain of hyping to just make it worthwhile to discuss in tech circles (removing the grifting on Linkedin). For everyone else, the most impact is the enormous amount of slop.
I paid for my computer. I'm paying for my internet connection and my electricity. I'm paying for my email service. Drop the free offering for any current AI service and see how much it is that transformative.
The VCR? Hrm. What part of the world are you from? No computer and no internet? Hrm.
> Drop the free offering for any current AI service and see how much it is that transformative.
It is currently being paid for by millions of companies in Microsoft products, and I can tell you that many companies are happy with it.
Microsoft honestly has done a tremendous job integrating it with some of their products, less so in others. (Excel is a disaster, Teams has been pretty good)
When each of those things came in, there was definitely some stumbling but I remember things being less ambiguous about being good than LLMs.
and none of them changed society more than the smartphone (starting with the iphone).
He’s on track to weigh 7.5 trillion pounds by the age of 10.
There’s nothing proving that this technology won’t be an S curve growth like every other before it.
This is some AI psychosis shit.
Explain how you think extrapolating off of 4-5 years of growth(I’m being generous here) means the development is asymptomatic instead of reaching a plateau like every other aspect of physics.
Are you one of those people who believes Anthropics claim that they will have the revenue of the entire United States yearly for their IPO?
The only one I saw upon review was
> Extrapolation of the extreme AI progress we have seen in the last few years.
I don’t think that’s a viable extrapolation hence my mockery about a 3 month old being trillions of pounds by the time they were 10 years old.
If there was another explanation please let me know so I can evaluate it.
You may claim it doesn't understand you, then I can say, fine... a machine that perfectly and indistinguishably imitates a response made by an intelligent being that understands such that a difference cannot be measurably detected.
I would say, the transistor was probably the biggest thing to happen in tech, full stop.
- The lighting of the match is mainly caused by the striking of the match, not by the presence of oxygen, even though both are necessary for lighting the match.
So a cause being necessary for an effect doesn't imply that its impact on the effect is very large.