usaar333

22 ms·

211.
by usaar333 3y ago
Yeah, but it's a comparison metric to the human ones, so is relevant to implying the GPT-4 set is "better".
212.
by usaar333 3y ago
I find it weird to compare GPT-4 to not expert humans (which this did - they seemed to have just sampled pitch decks for funded companies). GPT-4 generally will beat non-experts. What you want to see is a comparison to say pitch deck consu
213.
by usaar333 3y ago
Manhattan is some of the most expensive housing in the US. It's seeing significant pressure from relocations from other places (e.g. many San Franciscans have moved there)
214.
by usaar333 3y ago
Why would that affect anything? Reduced demand for house purchases should result in higher supply for rental housing as owners rent it out instead of sell.
215.
by usaar333 3y ago
Not necessarily. 1. It's entirely possible they could make more money monetizing the highly engaged users that use mobile apps through ads than those apps make directly. 2. Their cost structure is higher than the apps since they need t
216.
by usaar333 3y ago
#1 has high focus more due to impact than high probability. #2 doesn't seem talked about much at this point and seems to be pivoting more to #3. #2 never had much of a compelling argument given auditability. #3 gets mainly ignored due
217.
by usaar333 3y ago
If anything, the consequences seem minimal because it is so oversupplied.
218.
by usaar333 3y ago
> “In every previous automation threat, the automation was about automating the hard, dirty, repetitive jobs,” said Ethan Mollick, an associate professor at the University of Pennsylvania’s Wharton School of Business. What an odd stateme
219.
by usaar333 3y ago
I'd expect analysis explaining why firms are able to achieve "excess profit" now for such a strong claim to be believed.
220.
by usaar333 3y ago
I'm a bit confused by the productivity gain you are seeing. For me, it would be faster to just do the react work (assuming I'm in webstorm with its own fast autocomplete) than go back and forth into ChatGPT and explain what I wan
221.
by usaar333 3y ago
One was just posted: https://news.ycombinator.com/item?id=36152510
222.
by usaar333 3y ago
See this thread: https://news.ycombinator.com/item?id=36152510 It's not going to be barely useful, if at all, when you have high skill in a domain. It's quite useful when you don't.
223.
by usaar333 3y ago
I've personally found on net it negates ALL the benefits (on net) for areas I'm at least "modestly skilled" in. That is if I could only use GPT-4 vs. GPT-3, I'd take GPT-3 because the faster speed outweighs the hig
224.
by usaar333 3y ago
From a cursory read, they don't reveal the productivity difference between high skilled and low skilled workers ( https://www.nber.org/system/files/working_papers/w31161/w311... guessing this is conf
225.
by usaar333 3y ago
Full paper: https://www.nber.org/system/files/working_papers/w31161/w311... The most interesting thing to me is how little of a productivity boost you are seeing among higher skilled agents. Top quintil
226.
by usaar333 3y ago
> Personally I believe that GPT-4, and even GPT-3, are non-superintelligent AGI already, and as far as I know they haven't killed anyone at all. They aren't agentic. There's little worry a non-agentic AI can kill people. A
227.
by usaar333 3y ago
Metaculus puts the odds much higher - around 3%/year. https://www.metaculus.com/questions/11480/china-launches-inv... https://www.metaculus.com/questions/5320/chinese-annexation-...
228.
by usaar333 3y ago
Agreed failed is too strong, but they are doing worse than than their 2016 hype levels would suggest today. https://www.theverge.com/2016/9/18/12944506/lyft-self-drivin... <this was pretty absurd>
229.
by usaar333 3y ago
Agreed, but with Bing powered by GPT 4, there isn't that much advantage.
230.
by usaar333 3y ago
No, that's wrong. The answer is 4999/9999
231.
by usaar333 3y ago
> GPT4 should be better, but it is still limited beta and I haven't bothered joining. Ah my bad. Gpt4 via bing precise gets it correct: > A kilogram of feathers weighs more than a pound of bricks. A kilogram is a metric unit of
232.
by usaar333 3y ago
> TLDR: see this prompt and ChatGPT's response And wow, that's GPT4. I've had similar thoughts as you. It feels like amazing intelligence one day, but the next seems like a extremely good, but naive pattern matcher. I&#x27
233.
by usaar333 3y ago
Yes. The redwood forests in Del Norte County have enough rainfall to meet the temperate rainforest definition (55 inches of annual rain). And yes, you can get to the desert driving 3 hours east (albeit into Oregon)
234.
by usaar333 3y ago
I'd guess well under 1%. It's at 15% extremely useful for the users - maybe 5% of those are willing to pay? That still is huge though. 500k paying American users alone. Fwiw, I think people's experiences may vary to be willi
235.
by usaar333 3y ago
#1 is only true for remote work.
236.
by usaar333 3y ago
You might be getting lucky because that sequence is trained on (it's not very random). I repeated this experiment with bing and verified it could get your solution. But it failed on more random input: please could you multiply 926319
237.
by usaar333 3y ago
Pretty sure you can't get GPT-4 to do 8 digit multiplication with any prompt. For what it's worth, I'm not even sure if chain of thought provides much value to GPT-4. The RLHF it went through seems to have encouraged more log
238.
by usaar333 3y ago
AGI is just a poorly defined, moving target. Absent the goals constantly shifting, GPT-3 can be viewed as one, GPT-4 even more so. You can ask it questions about almost anything (at a broad level) and get an answer. That's what makes i
239.
by usaar333 3y ago
> Neeva failed because they didn’t understand distribution No, the team and its VCs fully understood distribution. Neeva never got to the point of having a compelling enough product where cost of user aq was below LTV for a large enough
240.
by usaar333 3y ago
Japan one seems wrong or at least wrongly explained. Japan controls Okinotorishima which is at 20 degrees north. But still impressive deductive reasoning.

More ›