Is AI the solution to search's problem?
lucaserb.substack.com
Is AI the solution to search's problem?
1–10 of 39 posts
Re: Is AI the solution to search's problem?
#2Using OpenAI to research and learn faster online
Re: Is AI the solution to search's problem?
#3Very interesting! And the bear-themed solution is cute! I would love to use such a tool to summarize and to ELI5 things like employment contracts or fine prints on a bank statement on the web. This tool might save folks a lot of time and energy.
Re: Is AI the solution to search's problem?
#4Using OpenAI to research and learn faster online
How reliable is the information that it returns?
Re: Is AI the solution to search's problem?
#5I think the next major leap will be when it accurately recommends real world books, videos, excerpts, papers, and products that naturally flow from your interaction.
Re: Is AI the solution to search's problem?
#6Maybe I’m an outliner. But I’m not buying it.
To me, it misses to analyze and address a fundamental problem of why google “got so bad”.
I believe this is mainly a problem of the top 5%. There is reason why google presents their results like they do. It’s not like they’re a bunch of idiots not knowing their craft.
Yes, for those of use around the internet since 1996, the results got worse. But since then the entire population got online. They’re just fine with the top results. They’re good enough. Google’s machine knows and sees this in the data.
I am sure there is value in summarizing and querying the text. But this will increase the convenience not necessarily the quality of results.
Re: Is AI the solution to search's problem?
#7First the LLM needs to know what it doesn't know, and it looks like except for the explicit limitations embedded by the developers it isn't capable of doing that. When I asked ChatGPT about its own model variants, it gave me wild answers for things that made sense but didn't exist.
"""
There are several variants of GPT-3 model, each fine-tuned for specific tasks or industries. Here are a few examples:
- GPT-3 "davinci" is fine-tuned for creative and imaginative writing tasks, such as poetry, short stories, and song lyrics.
- GPT-3 "curie" is fine-tuned for conversational and dialog generation.
- GPT-3 "babbage" is fine-tuned for code generation and language understanding.
- GPT-3 "einstein" is fine-tuned for question answering tasks.
- GPT-3 "jules" is fine-tuned for summarization.
- GPT-3 "parliament" is fine-tuned for legal text generation.
- GPT-3 "bronte" is fine-tuned for creative writing tasks like storytelling and fiction-writing.
"""
I'd almost assume "parliament" it's a leak of some sort if any of the other examples (except for davinci) were accurate.
Re: Is AI the solution to search's problem?
#8First the LLM needs to know what it doesn't know, and it looks like except for the explicit limitations embedded by the developers it isn't capable of doing that. When I asked ChatGPT about its own model variants, it gave me wild answers for things that made sense but didn't exist. """ There are several variants of GPT-3 model, each fine-tuned for specific tasks or industries. Here are a few examples: - GPT-3 "davinc…
Another possibility is that these code names do exist internally within OpenAI and chatgpt is aware of them.
Ah, i guess that's what you meant by leak. I think some of the others are accurate - babbage in particular is very accurate!
Re: Is AI the solution to search's problem?
#9Re: Is AI the solution to search's problem?
#10Google search from 2010-2015 was a great tool. Google search today is a sad shadow of its former self.