| ▲ | jstummbillig 6 days ago |
| Counterpoint: AI in its current form is democratizing and allowing exactly the not rich to be relatively more dangerous. So yeah, the rich might use it to get richer. But so can everyone* else. |
|
| ▲ | lm28469 6 days ago | parent | next [-] |
| > AI in its current form is democratizing and allowing exactly the not rich to be relatively more dangerous. Which part exactly ? The part where everyone pays 20+ a month to a few megacorps or the part where we willingly upload all our thoughts to a central server ? |
| |
| ▲ | whynotminot 6 days ago | parent | next [-] | | $20 a month for a nearly unlimited stream of high intelligence isn’t really undemocratic imo | | |
| ▲ | selfhoster11 6 days ago | parent | next [-] | | To call GPT-4o high intelligence, is aspirational (to put it more plainly: GPT-4o is such a bad model it's not worth paying for compared to what's out there). And yes, it is undemocratic - when was the last time you got a say over what the AI is allowed to do for you, let alone a say over any of the ideas for how to improve it? | | |
| ▲ | whynotminot 6 days ago | parent | next [-] | | 4o is pretty mid, you’re not wrong. Although for most things and most people it’s mostly fine. For my money (my actual money!) o3 is still the best model I’ve used. That is included in the $20 a month plan. | |
| ▲ | jstummbillig 6 days ago | parent | prev | next [-] | | $20 buys you o3 access. I have not run into limits for personal or professional information or research purposes. I am sure you can. | | |
| ▲ | selfhoster11 5 days ago | parent [-] | | Last time I had access to o3 on a $20 subscription, the usage limit was laughably low. Accessing it through an API is much better value - I get over 1000 interactions with a much more controllable system message, which is a lot better than ChatGPT Plus. | | |
| ▲ | whynotminot 4 days ago | parent [-] | | Was that a pretty long time ago? Because I use o3 daily and have never hit a limit. The only "limit" I have really with o3 is my patience -- it's a slow model for regular use. If I don't need its intelligence I'll use o4-mini or even 4o for speed, saving o3 for the prompts that really need it. These days the only model I find I get rate-limited on frustratingly quickly is 4.5. It's clear OpenAI does not really want you using that one, despite the fact that's very good (and probably very expensive for them)! Pretty underrated imo. |
|
| |
| ▲ | iamnotagenius 6 days ago | parent | prev [-] | | [dead] |
| |
| ▲ | lm28469 6 days ago | parent | prev | next [-] | | > stream of high intelligence I think you're overestimating what people use llms for. The only thing they're democratising is themselves | |
| ▲ | LtWorf 6 days ago | parent | prev | next [-] | | It's funny that you think they can't just raise the prices at will. (And by funny i mean really sad) | | |
| ▲ | whynotminot 6 days ago | parent [-] | | Do you think it’s a monopoly? Is it not a competitive market? I see many strong Western competitors along with an expanding array of high quality open source options out of China. | | |
| ▲ | LtWorf 4 days ago | parent [-] | | But the open source options out of china are forbidden because being spied by china is evil while being spied by USA is good… usual free market lovers dropping it whenever. | | |
| ▲ | whynotminot 4 days ago | parent [-] | | Forbidden by who? I can select Kimi as my model in Cursor right now at my American company. What are you even talking about | | |
|
|
| |
| ▲ | 6031769 6 days ago | parent | prev [-] | | "High intelligence"? Excuse me while I ROFL. | | |
| |
| ▲ | bartread 6 days ago | parent | prev [-] | | Yeah, exactly. Lots of people use AI, if they can afford the subscriptions, but it’s only the tech oligarchs who can control AI, including controlling access to it. Until you can run high quality models on affordable devices on your desk or in your hand the extent of the democratisation is much more limited than you might like. Perhaps OSS will come to the rescue here. (Aside: obviously free tiers are available but these are all hobbled in various ways: usage limits, data sharing/leakage, etc.) |
|
|
| ▲ | ndiddy 6 days ago | parent | prev | next [-] |
| If the AI industry achieves its short-term goals, instead of paying a human $100,000/year to do some desk job, companies will pay Microsoft/Google/OpenAI/whoever $20,000/year in API tokens and keep the extra money for themselves. To me, this doesn't seem like a way to reduce wealth inequality, it seems like a way to accelerate it. Sure, there's nothing inherent to AI that makes it cause wealth inequality. However, literally every innovation in human history that allows a single worker to generate more value has caused most of that extra value to get captured by the ownership class. I don't see how AI will be any different. |
| |
| ▲ | fragmede 6 days ago | parent | next [-] | | Even if it costs $100k/yr in GPU and RAM, the robot's preferable. It doesn't get sick, it doesn't have a family, it doesn't show up hungover, it doesn't sleep. It can be copy and pasted and duplicated and turned off as necessary. | |
| ▲ | jstummbillig 6 days ago | parent | prev | next [-] | | The point is that instead of having to complain about the ownership class, you can now be an owner more easily than before. AI enables people to do more on their own. You can build stuff that simply was not feasible before. You still have to do it though. If you prefer to be subject to the ownership class, I recommend being honest about why that is. | | |
| ▲ | ndiddy 6 days ago | parent | next [-] | | In that case you're still subject to the companies who have billions of dollars to spend on training frontier models and server farms full of GPUs. They're free to change their pricing or the quality of the services offered at will (see the recent Claude Code limit changes). Long-term, what you're describing would be a ton of underpaid "entrepreneurs" with no benefits all working away on whatever B2B SaaS web app while all the real money gets made by the AI services they subscribe to. It would be similar to how Uber has "democratized" taxis. | |
| ▲ | weakfish 6 days ago | parent | prev [-] | | I, for one, just don't really believe that there's enough market for every person to have their own software business. There's only so many ideas that are interesting enough to be sold. |
| |
| ▲ | buu700 6 days ago | parent | prev [-] | | Big companies may save money on salaries, which would have first-order effects on the job market — but you could say the same about cloud computing, or computers, or mechanical reapers. Now what about all the companies that previously couldn't afford to exist because someone with a good idea didn't have six+ figures of discretionary income to throw at it? If AI proves useful enough to effect sustained savings for Big Tech to the degree that you're suggesting, then the flip side of that coin is that it's effectively ZIRP on steroids for startups. The successful businesses which come out of that will ultimately have human labor needs of their own. Job market continues to trend upwards, only with lower concentration and greater overall resulting economic value in the form of more products and services. Or AI/robotics/etc. gets so good that eventually no one needs human labor, and every company trends toward just being a CEO + board + AI. In that world, unless we expect everyone and their grandma to become an entrepreneur, something to the effect of a UBI would be necessary to keep the world turning without major societal upheaval. On the other hand, AI skeptics can plan to sit back and eat popcorn while they watch the bigcos suffer the consequences of their mistakes and yield ground to startups. Either way, AI will have been a democratizing force. |
|
|
| ▲ | FirmwareBurner 6 days ago | parent | prev [-] |
| >So yeah, the rich might use it to get richer. But so can everyone else. N'ah as long as the AIs the everyone else has access to are heavily censored and lobotomized to prevent wrong think, while governments and corporations will have access to the raw unbiased data. |
| |
| ▲ | miltonlost 6 days ago | parent [-] | | The raw data will still be incredibly biased, and the AI will have its own biases on top of what's in the training. Grok will be racist no matter what. |
|
