4 ms·
> . Yes, it’s expanding the surveillance state, and yes, it’s destroying the education system, and yes, it’s being trained on copyrighted work without permissio
by notepad0x90 9mo ago
> . Yes, it’s expanding the surveillance state, and yes, it’s destroying the education system, and yes, it’s being trained on copyrighted work without permission, and yes, it’s being used to create lethal autonomous weapons systems that can identify, target, and kill without human input, but… I forget my point, but ultimately, I think you should embrace it.
What's your answer to this? How did it turn out for nuclear energy? If it wasn't for this sort of thinking we'd have nuclear power all over the world and climate issues would not have been as bad.
You should embrace it, because other countries will and yours will be left behind if you don't. That doesn't mean put up with "slop", but that also doesn't mean be hostile to anything labeled "AI" either. The tech is real, it is extremely valuable (I applaud your mental gymnastics if you think otherwise), but not as valuable as these CEOs want it to be or in the way they want to be.
On one hand you have clueless executives and randos trying to slap "AI" on everything and creating a mess. On the other extreme you have people who reject things just because it has auto-complete (LLMs :) ) as one of it's features. You're both wrong.
What Jensen Huang and other CEOs like Satya Nadella are saying about this mindless bandwagonning of "oh no, AI slop!!!" b.s. is true, but I think even they are too caught up in tech circles? Regular people don't to the most part feel this way, they only care about what the tool can do, not how it's doing it to the most part. But..people in tech largely influence how regular people are educated, informed,etc...
Look at the internet, how many "slop" sites were there early on? how much did it get dismissed because "all the internet is good for is <slop>"?
Forget everything else, just having an actual program.. that I can use for free/cheap.. on my computer.. that can do natural language processing well!!! that's insane!! Even in some of the sci-fi I've been rewatching in recent years, the "AI/Computer" in spaceships or whatever is nowhere near as good as chatgpt is today in terms of understanding what humans are saying.
I'm just calling for a bit of a perspective on things? Some are too close to things and looking under the hood too much, others are too far and looking at it from a distance. The AI stock valuation is of course ridiculous, as is the overhyped investments in this area, and the datacenter buildout madness. And like I said, there are tons of terrible attempts at using this tech (including windows copilot), but the extremes of hostility against AI I'm seeing is also concerning, and not because I care about this awesome tech (which I do), but you know.. the job market is rough and everything is already crappy.. I don't want to go through an AI market crash or whatever on top of other things, so I would really appreciate it on a personal level if the cause of any AI crash is meritocratic instead of hype and bandwagonning, that's all.
- ares623 9mo agoI wasn’t around at the time to argue against nuclear energy. I wasn’t old enough to argue against the internet. Plus to be fair to the ones who were, there was no prior tech that was anything like it to even make realistic guesses into what it would turn out to. I wasn’t old enough to argue against social media and the surveillance it brought. Now AI comes along. And I am old enough. And I am experienced enough in a similar space. And I have seen what similar technology have done and brought. And I have taken all that and my conscience and instinct tells me that AI is not a net good. Previous generations have failed us. But we make do with the world we find ourselves born into. I find it absurd that experienced engineers today look at AI and believe it will make their children’s lives better, when very recent history, history they themselves lived through, tells a very different story. All so they can open 20 PRs per day for their employers.
- notepad0x90 9mo agoWhether you were around or not is irrelevant, I wasn't around for some of that either. I brought it up so we can learn from the past instead of repeat those mistakes. > , there was no prior tech that was anything like it to even make realistic guesses into what it would turn out to. Same with LLMs. > AI is not a net good. You're falling into the same trap as previous generations when you do that. You won't actually end up fixing or improving the negative impacts of AI and your country/society will lose out big time in all sorts of ways. Tech doesn't make things bad, people do to the most part. Where AI is abused, it needs legislation, not resistance and you should know it is a LOT more nuanced than that. How is an LLM language translator for tourists being loped in the same bucket as LLMs being used to target people for assasination? Your lack of nuance is laziness, no political or ideological stand can justify that laziness. Nuclear energy had a lot of negatives, and people made the same types of arguments and outright banned it, next time you complain about climate change consider that your way of thinking might be part of the problem. Right now datacenter build outs are contributing to water scarcity for example, so instead of doing the hard and nuanced work of actually regulating and fixing that you oppose AI entirely. You do the easy thing, in the end we live in the real world and supply/demand economics rules, so your resistance is only performative at best, or catastrophic to the economy at worst. The latter part isn't for billionaries, and it isn't just for job markets, when the economy goes, all the climate change talk goes with it, all the EVs, green energy initiatives,etc... go, wars and crises increase, disease outbreaks increase. Doing the easy thing leads to this is my point, not that you need AI to prevent those. > I find it absurd that experienced engineers today look at AI and believe it will make their children’s lives better, when very recent history, history they themselves lived through, tells a very different story. 100x it would! although whose children depends on who regulates it first. My bet is China and the EU will regulate the crap out of it and extract the most value for themselves and future generations. AI is just solution, a tool, it isn't magic as you very well know. Companies have being using ML for surveillance for a long time. The US gov was using life of pattern analysis ML ten years ago to pick out assassination targets in Afghanistan and Pakistan. You have a fundamental lack of laws and a broken system of governance, don't take that out on tech.