Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Introducing deep research

openai.com

Post reply on HN
Print view View original

181–190 of 445 posts

Re: Introducing deep research

#181

Post by dutchbookmaker »

ldjkfkdsjnv wrote: post #76

So much cynicism and hate in these comments, especially as we are likely witnessing AGI come to life. Its still early, but it might be coming. Where is the excitement? This is an interesting time to be alive. HN has a huge cultural problem that makes this website almost irrelevant. All the interesting takes have moved to X/twitter

I would be more excited if it wasn't $200 a month to try.

I don't feel like OpenAI does a good job of getting me excited either.

Find the perfect snowboard? How can that idea get pitched and make the final cut for a $200 a month service? The NFL kicker example is also completely ridiculous.

The business and UX example seems interesting. Would love to see more.

Re: Introducing deep research

#182

Post by neynt »

pama wrote: post #156
rvnx wrote: post #67

Earlier quoted context omitted.

This is one of the actual questions: > In Greek mythology, who was Jason's maternal great-grandfather? https://www.google.com/search?q=In+Greek+mythology%2C+who+wa...

No it is not an actual question on this exam. From the paper: “To ensure question quality and integrity, we enforce strict submission criteria. Questions should be precise, unambiguous, solvable, and non-searchable , ensuring models cannot rely on memorization or simple retrieval methods. All submissions must be original work or non-trivial syntheses of published information, though contributions from unpublished res…

It's example #7 on https://lastexam.ai/

Re: Introducing deep research

#184

Post by pazimzadeh »

> In Nature journal's Scientific Reports conference proceedings from 2012, in the article that did not mention plasmons or plasmonics, what nano-compound is studied?

Aren't there more than one articles that did not mention plasmons or plasmonics in Scientific Reports in 2012?

Also, did they pay for access to all journal contents? that would be useful

Re: Introducing deep research

#186

Post by jmount »

I had no idea there was a market for "Compile a research report on how the retail industry has changed in the last 3 years. Use bullets and tables where necessary for clarity." I imagine reading such a result is pure torture.

Re: Introducing deep research

#187

Post by nicce »

pazimzadeh wrote: post #184

> In Nature journal's Scientific Reports conference proceedings from 2012, in the article that did not mention plasmons or plasmonics, what nano-compound is studied? Aren't there more than one articles that did not mention plasmons or plasmonics in Scientific Reports in 2012? Also, did they pay for access to all journal contents? that would be useful

Maybe that is the only one with open access

Re: Introducing deep research

#188

Post by random_cynic »

Bjorkbat wrote: post #174

Actually sounds pretty cool, but the graph on expert level tasks is confusing my expectations. Saying it has a pass rate of less than 20% sounds a lot like saying this thing is wrong most of the time. Granted, these strike me as difficult tasks and I’d likely ask it to do far simpler things, but I’m not really sure what to expect from looking at these graphs. Ah, but the fact that it bothers to cite its sources is a…

I think that's mostly because of the access to information it has. Much of the highly useful information is not on the public internet or shows up on search engines, only domain experts know about them. Also, the websites may be paywalled or gated by login. So a better comparison would be if the models had the same level of access as an expert.

Re: Introducing deep research

#189

Post by blackeyeblitzar »

lompad wrote: post #25
blackeyeblitzar wrote: post #14

Earlier quoted context omitted.

I’m not sure I understand what you mean by “the button”. If you’re comparing this to DeepSeek’s copying, it’s not really the same thing right? DeepSeek essentially stole intellectual property by violating OpenAI’s terms of service. As I understand it, this is a copy of Google’s Deep Research

Deepseek proved that there is no moat. Thus no path to profitability for openai, anthropic & co. Stealing from thieves is fine by me. Sama was the one claiming that all information could be used to train LLMs, without permisdion of the copyright holders. Now the same is being done to openai. Well, too bad.

> Stealing from thieves is fine by me. Sama was the one claiming that all information could be used to train LLMs, without permisdion of the copyright holders.

OpenAI and other LLMs scraping the internet is probably covered under fair use. DeepSeek’s violation of OpenAI’s terms is pretty clearly a violation of their terms and not legal.

Re: Introducing deep research

#190

Post by Davidzheng »

itkovian_ wrote: post #122
YmiYugy wrote: post #24

If I understood the graphs correctly, it only achieves 20% pass rate on their internal tests. So I have to wait 30min and pay a lot of money just to sift through walls of most likely incorrect text? Unless the possibility of hallucinations is negligible, this is just way too much content to review at once. The process probably needs to be a lot more iterative.

Here's an example of the type of question it is acheiving 20% on; The set of natural transformations between two functors F,G ⁣:C→DF,G:C→D can be expressed as the end Nat(F,G)≅∫AHomD(F(A),G(A)). Nat(F,G)≅∫A HomD (F(A),G(A)). Define set of natural cotransformations from FF to GG to be the coend CoNat(F,G)≅∫AHomD(F(A),G(A)). CoNat(F,G)≅∫AHomD (F(A),G(A)). Let: - F=B∙(Σ4)∗/F=B∙ (Σ4 )∗/ be the under ∞∞-category of the ne…

btw isn't this question at least really badly worded (and maybe incorrect?) the definitions they give for F and G are categories not functors... (and both categories are in fact one object with contractible space of morphisms...)
Post reply on HN