What can GPT-3 do that is useful? I can understand that it outputs text based on a prompt and some input but I don't understand how that can be leveraged to do useful things. I can ask it trivia questions but is it better than doing a Wikipedia search? Is it possible to give it a prompt of some scientific paper say and write an article about it? Or will it just generate nonsense?
At the moment it feels like the only use-cases are:
- toys/games: AI Dungeon, fun chatbots, generally playing around with generating text in a certain style
- humour: generating jokes/memes
- deception: trolling/disinfo campaigns/spam/gaming advertisement market by creating garbage content.
I guess there will be use cases where a human operator can use it to simplify their job (or lower the skill level required to do a job) by curating and editing generated content instead of writing it themselves. I'm thinking things like simple writing jobs where quality isn't that important: social media posts, newsletters?
I like the idea of using it as a tool to mitigate writers block. Say you're writing a school paper and you've done your research but you can't seem to keep your train of thought flowing. With gpt3 you could have it generate a sentence prompt based on your previous writing, or maybe you have it write your conclusion for you that summarizes your thoughts from the paper. There are lots of possibilities in that arena.
Even as far back as Eliza, there were reports of people finding that talking to the chatbot was therapeutic, and people would spill their guts out to it for hours.
With GPT-3's conversation having much more verisimilitude, the perceived therapeutic value should be much greater.
Many lonely, hurting people want someone to talk to, and for some knowing that they're talking to a machine increases their comfort in revealing personal details to it.
"Conversation as a service" could be a very desirable product for many.
It's somewhat impressive but it didn't do what he wanted it to. It didn't list temperatures, instead it listed weather descriptions. And to get it to do this he had to write a full template. We also don't get to see if this was the first attempt at doing this or can the model generate code like this reliably.
You are pushing this idea so hard and blasting this thread with copy/paste comments. It's obvious you're excited about it and that's great but there is a lot that isn't shown and that's where the real work is being done.
Just because it looks like magic, doesn't mean there isn't someone pulling some strings somewhere else to aid the illusion.
>You are pushing this idea so hard and blasting this thread with copy/paste comments.
I'm sorry but this comment is absurd. I'm assuming you are insinuating that I am astroturfing for OpenAI which is against guidelines. Not only that but in fact none of my comments are copy pasted so its doubly ridiculous.
>Just because it looks like magic
No is saying it's magic but the thread is full of people saying: "Uhh my random prompt got bad results, this is just hype, blah blah blah..." People looking for excuses to trash the model instead of seeing what it could mean for the industry going forward. None of this is married to OpenAi either, there are plenty of groups replicating GPT and they will likely have similar capabilities.
What if you ask it a serious question you are genuinely interested in getting an answer to and it responds with a real clue? They say it seems like going to reach the human level, doesn't this mean it will NOT "just generate nonsense"? Humans use to generate nonsense but it often makes at least some sense.
People need to look WAY beyond generating text here. There are lots of demos on Twitter already of people generating fully functioning code in various languages with english inputs.
I tried the recommendation engine with some prompts that are fairly simple, but would require a deeper understanding than just a keyword search to get right (e.g. "books where magic is secretly technology"). The recommendation engine does not seem to be doing significantly better than a keyword search would. I would probably have more success with raw Google.
This is the sort of thing that dampens the hype for me a bit. I keep assuming that the demonstrations are not cherrypicking examples, but it's kind of hard to believe I'm just uniquely good at picking problem cases.
Actually I like the results a lot. How does the logic behind it work, is it just the output of GPT-3? Do you need to parse it somehow? If it is just GPT-3 its kind of amazing how good the results are, since it was never trained for that. A bit scary to be honest.
It's mostly just GPT-3. You do need to parse it, as far as GPT-3 is concerned it's just doing text in/text out. I'm doing some ranking logic behind the scenes to make the results more consistently reliable, but you could directly ask the model for recommendations. GPT-3 has a lot of knowledge of the real world from having read so much of the internet.
Wow kind of amazing. Don't think a human would give me much better results. It would be interesting to know how fast OpenAi is able to retrain / update the model based on new data.
