Search is the primary means by which we navigate and access digital information. It lies at the heart of the modern internet experience.
Today’s large language models can read and write with a level of sophistication that a few years ago would have seemed inconceivable. This will have profound implications for how we search.
In the wake of ChatGPT, one reconceptualization of search that has gotten a lot of attention is the idea of conversational search. Why enter a query and get back a long list of links (the current Google experience) if you could instead have a dynamic conversation with an AI agent in order to find what you are looking for?
Conversational search has a bright future. One major challenge needs to be resolved, though, before it is ready for primetime: accuracy. Conversational LLMs are not reliably accurate; they occasionally share factually untrue information with total confidence. OpenAI CEO Sam Altman himself recently cautioned: “It’s a mistake to be relying on ChatGPT for anything important right now.” Most users will not accept a search application that is accurate 95% or even 99% of the time. Addressing this issue in a scalable and robust way will be one of the primary challenges facing search innovators in 2023.
You.com, Character.AI, Metaphor and Perplexity are among the wave of promising young startups looking to take on Google and reinvent consumer search with LLMs and conversational interfaces.
But consumer internet search is not the only type of search that LLMs will transform.
Enterprise search—the way that organizations search and retrieve private internal data—is likewise on the cusp of a new golden age. Thanks to large-scale vectorization, LLMs enable true semantic search for the first time: the ability to index and access information based on underlying concepts and context rather than simple keywords. This will make enterprise search vastly more powerful and productive.
Startups like Hebbia and Glean are leading the charge to transform enterprise search using large language models.
And the opportunities for next-generation search extend beyond text. Recent advances in AI open up whole new possibilities in multimodal search: that is, the ability to query and retrieve information across data modalities.
Given that it accounts for ~80% of all data on the internet, no modality represents a bigger opportunity than video. Imagine being able to search effortlessly and precisely for a particular moment, individual, concept or action within a video. Twelve Labs is one startup building a multimodal AI platform to enable nuanced video search and understanding.
Search has changed surprisingly little since Google’s ascendance during the dot-com era. Next year, thanks to large language models, this will begin to change dramatically.
If you enjoyed this market, please check out the other 9! https://manifold.markets/group/forbes-2023-ai-predictions
This market is from Rob Toews' annual AI predictions at Forbes magazine. This market will resolve based on Rob's own self-assessed score for these predictions when he publishes his retrospective on them at the end of the year.
Since Rob resolved and graded his 2022 predictions before the end of 2022, I am setting the close date ahead of the end of the year, to (try to) avoid a situation where he posts the resolutions before the market closes. In the event that his resolution post falls in 2024, my apologies in advance. If he hasn't posted resolutions at all by February 1, 2024, I will do my best to resolve them personally, and set N/A for any questions that I can't determine with outside source data.
-----
Edit 2023-07-05: Last year Rob used "Right-ish" to grade some of his predictions. In cases of a similar "Right-ish" (or "Wrong-ish") answer this year, I will resolve to 75% PROB or 25% PROB, respectively. This will apply for similar language too ("mostly right", "partial credit", "in the right direction"). If he says something like "hard to say" or "some right, some wrong", or anything else that feels like a cop-out or 50% answer, I will just call that N/A.
1,000
3.00
@MattCWilson lmao what a fraud. I guess it's worth losing a thousand mana to learn the lesson of never trusting this grifter.
edit: no hate to you matt, you resolved it fairly
@Kurgan Nah man, none taken! 💯 with you. Lost a bunch of mana on these myself; agree that the lesson here is "question the predictions of pundits." 😂
@MattCWilson Good lord what a moron.
I am getting to the point that I'd like some kind of plug-in that notes the calibration of any public figure. It'd be nice if anyone making predictions on e.g. twitter was shamed by the community for being demonstrably and frequently wrong, y'know?
My earliest memories of using the internet are in the late 2000s, and I clearly remember using web portals alongside Google. Google was just not considered to reliable enough at that time. I also remember how groundbreaking being able to search your emails in gmail was. That was only like 10 years ago. Pretty insane that Rob thinks that the change this year is greater than all the changes between the early 2000s and now.
@AndrewHartman I believe if we continue to promote MM as the definitive home of online accountability, Twitter will eventually catch on and can do the shaming for us 😉
@MattCWilson this resolves Yesssssssssssss baby 🥰
@HenriThunberg This is bullshit, but I guess I should have expected this bullshit to happen, given the description.
Oh shit, I didn't realize/forgot this got locked in last night!
@MattCWilson would you be interested in opening it up a for a pre-specified time period (e.g. 24 hours on Monday) so that people can place their "final bets"? I think that could be an interesting last squeeze out of this market that has had surprising volatility throughout the year despite little changes in the actual underlying facts.
@HenriThunberg Ahh, hadn’t seen this when you posted it. But even then, I think probably not, since the market was open all year a last-minute change would be unexpected.
@MattCWilson Easy to say now, but I would have wanted to push it up at least to 60% if I had the chance and the liquidity. Also Gemini Pro release seems like something that easily could have changed things (even if in this case I don't think it did).
I'd also be curious about your thoughts on just allowing people to news trade on it instead. My conclusion has been that it's basically just adding a different extra layer to the already existing market, rather than having any implications on the existing bets.
Anyway, thank for a well-run set of markets! Do you plan to post the 2024 ones when (?) they come out?
I think y'all are underestimating how much Google has become better over the last 20 years. I still remember having to look at the second page of google results, and natural language questions such as "how tall is Donald Trump" not working well at all. You had to search keywords. But hey, I'm too young to remember everything so maybe I'm wrong.
@Shump I agree with your statement if we were mostly going by the title. However, his description doesn't have the vibe that this would be enough for a NO at all, IMO. 🤷♂️
@HenriThunberg Yeah that's why I only bet a small amount. The generative AI hype is strong and idk about Rob but since he writes about AI he is probably a "hyper". But I really don't think this should resolve yes.
@Shump Looking at the specific types of search mentioned, only converasational search has advanced significantly this year. But it also advanced significantly in the last 20 years, from nonexistent to being able to get answers to conversational-style questions. Enterprise search barely advanced this year, but 20 years ago it simply didn't exist. The infrastructure to allow that was not there. Video search is still not a thing.
@Shump
"Search has changed surprisingly little since Google’s ascendance during the dot-com era. Next year, thanks to large language models, this will begin to change dramatically."
I agree with the AI hype aspect being fairly strong, I also believe that the above part of the column gives a lot of wiggle room if he feels he needs it.
@HenriThunberg Yeah this checks out in my anecdotal experience as well. Seems to me that ChatGPT is a lot better at giving descriptive answers that you eventually learn to trust. Google asks you to pick something to look at the then scroll to find the answer and you’re still not super sure to trust it which compared to GPT is slow and clunky.
What do you mean by voice ChatGPT?
@gaurav Voice for interacting with ChatGPT https://openai.com/blog/chatgpt-can-now-see-hear-and-speak
@HenriThunberg Interesting. I still don't think it's useful if you know how to internet search quickly.