poormathskills
Re:It seems that OpenAI is scraping [certificate transparency] logs
on Monday December 15, 2025 @11:10AM (#46276367)
Is it still “scraping” when the purpose of these transparency logs is to be used for this purpose?
OpenAI has never compared their models to models from other labs in their blog post. Open literally any past model launch post to see that.
For a minor version update (5.1 -> 5.2) that's a way bigger improvement than I would have guessed.
Re:OpenAI to buy AI startup from Jony Ive
on Wednesday May 21, 2025 @01:32PM (#44053903)
OpenAI does literally anything
“This is a sign of OpenAI’s weakness”
I think this is the third time I’ve seen this exact comment at the top of a HN post about an OpenAI announcement.There is a weird amount of emotional investment in not wanting OpenAI to win.
Personally, I am just excited to see what the device looks like. The prototype must be good to justify this valuation.
Re:Evolving OpenAI's Structure
on Monday May 05, 2025 @06:16PM (#43899976)
>I'm very rarely unique in my behaviour
I cannot stress this enough: if you know what Deepseek, Claude, Mistral, and Perplexity are, you are not a typical consumer.
Arguably, if you have used a single one of those brands you are not a typical consumer.
The vast majority of people have used ChatGPT and nothing else, except maybe clicking on Gemini or Meta AI by accident.
Re:GPT-4.1 in the API
on Monday April 14, 2025 @04:49PM (#43686168)
Who has a (publicly released) model that is SOTA is constantly changing. It’s more interesting to see who is driving the innovation in the field, and right now that is pretty clearly OpenAI (GPT-3, first multi-modal model, first reasoning model, ect).
Re:GPT-4.1 in the API
on Monday April 14, 2025 @04:47PM (#43686138)
Right. Those labs aren’t leading the industry.
Re:GPT-4.1 in the API
on Monday April 14, 2025 @01:32PM (#43683859)
Go look at their past blog posts. OpenAI only ever benchmarks against their own models.
Re:GPT-4.1 in the API
on Monday April 14, 2025 @01:31PM (#43683840)
Go look at their past blog posts. OpenAI only ever benchmarks against their own models.
This is pretty common across industries. The leader doesn’t compare themselves to the competition.
Re:Trump temporarily drops tariffs to 10% for most countries
on Wednesday April 09, 2025 @03:34PM (#43636714)
Trump tweeted "it's a good time to buy" right before the tariff drop announcement.
Re:New funding to build towards AGI
on Monday March 31, 2025 @06:46PM (#43540791)
Article says ChatGPT, so that doesn't include users through the API.
Re:New funding to build towards AGI
on Monday March 31, 2025 @06:42PM (#43540758)
There have been news reports for this funding round for months, e.g. this one from January https://www.wsj.com/tech/ai/openaiin-talks-for-huge-investme...
Re:New funding to build towards AGI
on Monday March 31, 2025 @06:41PM (#43540745)
The most surprising thing is that OpenAI reported 400 million weekly actives only last month: https://www.reuters.com/technology/artificial-intelligence/o...
Re:GPT-4.5 is #1 on Chatbot Arena (LMSYS) in all categories
on Monday March 03, 2025 @02:34PM (#43245750)
I’m surprised it topped the reasoning models for code generation and hard prompts. The style control results are also impressive: https://twitter.com/lmarena_ai/status/1896590154871210154
Re:Airbnb raises violent crime rates in cities as residents are pushed...
on Wednesday July 21, 2021 @11:35PM (#27915045)
I agree that it seems to be a pretty poor paper.
>They even acknowledge that the correlation could be due to a factor that isn't being accounted for.
Related to this, they're performing a regression between Airbnb prevalence and crime, but they only control for a single variable: income[0]. They look at others in a robustness check, but a single control variable practically screams p-hacking.
They also don't address the fact that both Airbnb prevalence and crime are nonstationary[1]. Regressing two nonstationary time series results in a nonsense coefficient[2]. Two totally unrelated time series will have a high coefficient if both exhibit consistent trends.
[0] "We report the results based on using income as the main tract-level control variable"
[1] They are consistent trends over time, see https://www.investopedia.com/articles/trading/07/stationary....
[2] https://stats.stackexchange.com/questions/94723/using-non-st...