The fact that transformers were immediately applied to pretty successfully doing text to text with ChatGPT, Text to Image with Stable Diffusion, and audio to text with whisper. All blowing away any previous attempts in all of these modalities, and showing that you can pretty easily go between the modalities is proof to me that this is a big deal. I see how fast people are able to use LoRas to make the models do different things, I see how excited people are to use them and their sustained interest and experimentation, and I can not possibly over state how confident I am they will only grow in impactfulness as they are combined and tweaked and experimented with.
AI demand is already shrinking
91–100 of 116 posts
Re: AI demand is already shrinking
#92...because the technology cannot yet live up to the insane hype about and around it ... fueled by over-enthusiastic promoters selling snake-oil. There are still quite a few challenges to making LLMs work reliably. For example, see https://huyenchip.com/2023/08/16/llm-research-open-challenge... . Until AI researchers and practitioners find ways to overcome those challenges -- possibly requiring new breakthroughs -- th…
>There are still quite a few challenges to making LLMs work reliably.
I'm not unsold that AI will continue to improve but I'd be weary of this claim. LLM's reliably do what they are supposed to do and it appears they're reaching an asymptote in capability. If what you're wanting is a truthful chatbot, you're probably barking up the wrong tree with LLMs.
Do LLMs have utility? There's enough people on HN to convince me there are legitimate uses for them. However, many of the people using them are also the people desperate to legitimize their utility, so it's honestly hard to know. I personally have not found any application for LLMs in my life and would discourage anyone working on anything remotely mission critical from using them for code generation. But I've definitely heard of a few real world uses.
Re: AI demand is already shrinking
#93It's weird to me that some people just don't see the oncoming train. AI is here to stay. Use cases are still being worked out, but it's here to stay. Reminds me of the Astound application around 1995-1996. It was a pretty major PowerPoint competitor and I used it a lot. For whatever reason, they only had a Windows 3.1 version. So I called them asking when the Windows 95 version was coming out. They told me they didn'…
AI has been "the next big thing" since the 30's. The problem with AI is they still haven't found out how to actually deliver on that second word - intelligence. Until then, it's nothing more than an interesting toy.
Re: AI demand is already shrinking
#94It's weird to me that some people just don't see the oncoming train. AI is here to stay. Use cases are still being worked out, but it's here to stay. Reminds me of the Astound application around 1995-1996. It was a pretty major PowerPoint competitor and I used it a lot. For whatever reason, they only had a Windows 3.1 version. So I called them asking when the Windows 95 version was coming out. They told me they didn'…
Funny, because AI has seen plenty of applications before ChatGPT, yet "the oncoming train" in most peoples heads is "this is going to completely replace artists, programmers, etc" This is what is laughable, and has been from the very beginning. Will we _continue_ to use AI to try and find novel drug treatments? Yes. Will chatbots still be the first thing a customer support flow uses (and will it still suck)? Yes.
I strongly disagree. LLMs made AI accessible to the masses in a way that previous tools couldn’t.
Instead of having to collect a bunch of data, clean it, and train on it, you can simply write plain, human prompts. That’s extremely powerful.
Re: AI demand is already shrinking
#95It really looks like ChatGPT suffered from some form of "initial over hype syndrome". I'm not underestimating it, actually I think the application of transformers to do what ChatGPT is capable of doing is something really impressive, novel and caught the world by surprise. I'm sure we'll find many equally impressively good applications for it. But people eventually found its limitations. And did it quite fast. People…
This is what i thought, until i stumbled across ReAct prompts. The tldr is you prompt the llm to give its thought, provide it with guides for actions to take,(you do this through code) then give the answer back to the llm, itll continue in this loop. for the lennon questions
> Me: What was the moon phase when john lennon was killed?
> GPT: Thought: I need to find out the moon phase on the date when John Lennon was killed. Action: google: Moon phase on December 8, 1980 PAUSE
I didnt code this I just did the search my self following is the google result
> ME: The Moon phase for December 8th, 1980 is a Waxing Crescent phase.
> GPT: Observation: The Moon phase for December 8th, 1980 was a Waxing Crescent phase. Answer: The Moon phase when John Lennon was killed on December 8th, 1980 was a Waxing Crescent phase.
these iterative thoughts and actions really sold me on it. theres papers that go more indepth
Re: AI demand is already shrinking
#96...because the technology cannot yet live up to the insane hype about and around it ... fueled by over-enthusiastic promoters selling snake-oil. There are still quite a few challenges to making LLMs work reliably. For example, see https://huyenchip.com/2023/08/16/llm-research-open-challenge... . Until AI researchers and practitioners find ways to overcome those challenges -- possibly requiring new breakthroughs -- th…
>There are still quite a few challenges to making LLMs work reliably. I'm not unsold that AI will continue to improve but I'd be weary of this claim. LLM's reliably do what they are supposed to do and it appears they're reaching an asymptote in capability. If what you're wanting is a truthful chatbot, you're probably barking up the wrong tree with LLMs. Do LLMs have utility? There's enough people on HN to convince me…
I agree, it may not be possible to overcome those challenges with LLM's specifically. We could well end up needing something other than LLMs, or new kinds of AIs that use LLMs. No one yet knows.
Re: AI demand is already shrinking
#97[dead]
Re: AI demand is already shrinking
#98Typical Gartner Hype Cycle pattern, just faster than normal.
Re: AI demand is already shrinking
#99Earlier quoted context omitted.
It’s suffering from extreme overhype pushed by OpenAI. They wanted people to think the human-led world was over and they were first to market on creating the new overlords. This soured them on a lot of tech people and ensured whatever they built could never live up. Still a great product and I use a chatbot everyday for work, but it’s just that a chatbot.
> extreme overhype It's hard to gauge hype except retrospectively. We will know in 4-6 years if it was over hyped.
I don't know. Historically, it seems pretty easy to spot extreme overhype in the moment, without the need for years of hindsight. The most obvious sign that this was excessively overhyped was when people were seriously arguing about whether AGI is right around the corner.
Re: AI demand is already shrinking
#100It would make perfect sense to me if the "demand" is more of "excitement about the idea" than "desire to make this into a long-term product." I think the actual introduction of these technologies into our lives is going to be a lot slower and more constrained. Every business would love to get rid of customer service agents and replace them with AI, for example. But no business wants to take a CS agent with an 85% acc…
> But no business wants to take a CS agent with an 85% accuracy/successful call record and replace it with an AI that's at 40-50%
I'm not so sure about that. Businesses generally had no problem outsourcing customer service to call centers even though doing that came with a very steep reduction in accuracy and successful resolutions. Those monetary savings are very tempting.