Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Sol-
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
91.
▲
by
Sol-
8mo ago
What's the copyright situation with music models? I was always under the impression that it will be political pressure and lawsuits by the existing labels that will be the biggest hindrance in the adoption of AI music models, rather th
92.
▲
by
Sol-
8mo ago
Seems like policy ripe with unintended side effects. At the very least, it'll likely raise prices for consumers because the companies aren't allowed to manage their inventory as efficiently as they wish. Now of course this might b
93.
▲
by
Sol-
8mo ago
Cheap clothing is a civilizational achievement and good for human welfare. So carbon emissions are bad, but then we should price carbon and not micromanage clothing inventory.
94.
▲
by
Sol-
8mo ago
That was my immediate reaction as well. These things are never more than a storm in a teapot.
95.
▲
by
Sol-
8mo ago
Because progress and growth makes us wealthier and happier? It's pretty simple. People say "Oh, but GDP isn't everything" - but it's correlated with almost everything good, so might as well be.
96.
▲
by
Sol-
8mo ago
I don't think this is the case, according to the docs, right? The effort level will use fewer tokens, but the independent fast mode just somehow seems to use some higher priority infrastructure to serve your requests.
97.
▲
by
Sol-
8mo ago
Dunno, it's probably less energy efficient than a human brain, but being able to turn electricity into intelligence is pretty amazing. RAM and power generation are engineering problems to be solved for civilization to benefit from this
98.
▲
by
Sol-
8mo ago
With stuff like this, might be that all the infra build-out is insufficient. Inference demand will go up like crazy.
99.
▲
by
Sol-
8mo ago
Is it their version of virtual AI employees that some startups were previously getting into, plus on-site support by FDEs and such?
100.
▲
by
Sol-
8mo ago
I really like that they establish it as their niche, and given their corporate focus it's also easy for them to take that stance. They have less to lose from forgoing advertising than ChatGPT with its relatively larger consumer focus.
101.
▲
by
Sol-
8mo ago
Dunno, I want to agree, but at the same time it's spoken like someone to whom these experiences and human relationship come easily. There are many people out there who, for some reason (anxiety, etc.), cannot easily access this part of
102.
▲
by
Sol-
8mo ago
I also find the implications for this for AGI interesting. If very compute-intensive reasoning leads to very powerful AI, the world might remain the same for at least a few years even after the breakthrough because the inference compute sim
103.
▲
by
Sol-
9mo ago
Do they assume that the current state of our institutions is normatively correct? AI progress will come and have manifold benefits, therefore we shouldn't really restrict it too much. If the institutions cannot handle that, they will h
104.
▲
by
Sol-
9mo ago
I think it's still worthwhile, though. AI, given its current trajectory, will be able to help immensely with science and engineering challenges. Degrowth isn't a recipe for sustainable reduction of CO2 emissions.
105.
▲
by
Sol-
9mo ago
His tone and sucking up to his authoritarian government will probably only serve to negatively polarize Europe against Cloudflare, even if he might have a point of the substance itself.
106.
▲
by
Sol-
9mo ago
The shape of the politics changed, though. From civil rights, questioning authority and cypherpunk, which inherently has a libertarian bent, there's now much more identity politics and social justice / grievance culture with only
107.
▲
by
Sol-
10mo ago
Then just continue to take them? The article just once seems to mention side effects and these were suspected to be related to rapid weight loss and less to the drug itself. Pointless human interest story with some rent-a-quote expert sprin
108.
▲
by
Sol-
10mo ago
> TBD Lab’s researchers have come to view many Meta executives as interested only in improving the social media business That cannot have been a surprise to anyone joining.
109.
▲
by
Sol-
10mo ago
No, I think apparently it was used in the reinforcement learning step somehow to influence the model's final fine-tuning. At least how I understood it. The actual system prompt from Anthropic is shorter and also public on their website
110.
▲
by
Sol-
11mo ago
Is this a satirical comment in the sense that it reads so AI generated?
111.
▲
by
Sol-
11mo ago
Why? These models just leapfrog each other as time advances. One month Gemini is on top, then ChatGPT, then Anthropic. Not sure why everyone gets FOMO whenever a new version gets released.
112.
▲
by
Sol-
11mo ago
I think in the end they will just pay off copyright holders. The German GEMA is mostly interested in rent-seeking through whatever means available, it's basically the whole point of the organization. They'll easily be paid off onc
113.
▲
by
Sol-
11mo ago
I remember the book "Algorithms to Live By" actually also uses finding a life partner as a fun, if perhaps unrealistic, example of applying the secretary problem. As far as I remember, it jokingly assumes that one's active da
114.
▲
by
Sol-
11mo ago
I do like that they notice the fact that automation and plummeting wages do not automatically mean immiseration for the population. I've read so many uninformed online discussions along the lines of "If no one has jobs anymore, wh
115.
▲
by
Sol-
1y ago
The amount of drama about AI based upscaling seems disproportionate. I know framing it in terms of AI and hallucinated pixels makes it sound unnatural, but graphics rendering works with so many hacks and approximations. Even without modern
116.
▲
by
Sol-
1y ago
Nowadays I always wonder to what extent such deals are actually driven by market considerations and to what extent it's catering to the Trump administration. Token investments into this state enterprise named Intel seems to be a practi
117.
▲
by
Sol-
1y ago
People are clueless about economics, so such polls are irrelevant. It can still have been a bad deal, of course, who knows, given the specifics of trade deals with Trump are ever changing. Ultimately it just reflects the EU's largely s
118.
▲
by
Sol-
1y ago
> all within the 4.5-hour competition time limit Both OpenAI and Google pointed this out, but does that matter a lot? They could have spun up a million parallel reasoning processes to search for a proof that checks out - though of course
119.
▲
by
Sol-
1y ago
How much demand for software is there, though? I don't buy the argument that the cake will grow faster than jobs are devalued. On the bright side, prices might collapse accordingly and we'll end up in some post scarcity world. No
120.
▲
by
Sol-
1y ago
Is that not just how scaling goes? It generally feels like the top models are mostly interchangeable and the one that came out at time t+1 will be better than earlier models from time t. Grok 4 has probably been training when O3 was release
More ›