willy_k

30 ms·

121.
by willy_k 1y ago
Ironically an AI written article.
122.
by willy_k 1y ago
The router seems to only apply to the ChatGPT version, not the API, so it’s not really anything new. Gemini already has functionally dynamic reasoning effort.
123.
by willy_k 1y ago
Yes they do, if the model size / vram requirement keeps shrinking for a given performance target, like has been happening, then it gets cheaper to run X level of model.
124.
by willy_k 1y ago
For API users, yes, but for the average person with a subscription or using the free tier it’s the inverse.
125.
by willy_k 1y ago
That doesn’t really matter for the point at hand, which is that Europe’s overregulation stifles innovation and international competition. Unless you’re arguing that the reason Europe hasn’t produced a SOTA LLM is because it’s not a net be
126.
by willy_k 1y ago
FWIW the fair comparison for Gemini 2.5 Pro would be o4-mini. That being said I’ve also found that Gemini is way better at getting it right on the first or second try and responding to corrections.
127.
by willy_k 1y ago
You don’t have to be happy with it. If Microsoft already has all your data it’s much easier to trust them for AI than to add another entity into the equation, especially OpenAI.
128.
by willy_k 1y ago
o3 is known to use web search to reference documents directly and be perhaps a bit overly diligent when it comes to double checking itself. Of course it can still make mistakes, but for something like this that’s a lot less likely than with
129.
by willy_k 1y ago
I’m as much a fan of thoughtless AI forwarding as the next HNer, but this seems like a fair situation for it: https://chatgpt.com/share/68199e8e-5ad4-8012-8026-09fa353b60...
130.
by willy_k 1y ago
Just pointing out that that is the case for Christianity too (not religion in general because others that I am aware of require some sort of actions be taken, although some would say that true acceptance necessarily leads to actions so ther
131.
by willy_k 1y ago
Maybe some of the companies marketing it, of course there are people looking to exploit it just like in nearly every other industry. But the concept itself isn’t some ADHD specific thing, it’s just using technology to exploit a quirk in hum
132.
by willy_k 1y ago
The best I’ve found for focus is simple caffeine. Amphetamines are good for motivation but give me too much motivation to do things other than what I need to be doing. Methylphenidate is maybe a bit better but still a similar story. They al
133.
by willy_k 1y ago
You’re not dumb, it’s an incredibly complex topic with convoluted and contradictory messaging everywhere. The way that I’ve learned to think about it is that the brain is made up neurons, and they perform specific functions, technically ind
134.
by willy_k 1y ago
You could just as (if not more) easily gain clarity into what you need to fix and how, and increase your ability to act on those insights.
135.
by willy_k 1y ago
Nicotine doesn’t really have any good MOAs. Inhalation is very short lasting, addictive, and gross and unhealthy. Pouches are harsh and still don’t last very long, gum (especially Lucy brand) is better but still has the same issues. And pat
136.
by willy_k 1y ago
I believe the point is that in other countries they won’t rifle through your bag to verify whether or not you have brought apples. I’m not familiar with Australian customs though so I could be mistaken.
137.
by willy_k 1y ago
That’s exactly what it is, and they said as much repeatedly while campaigning. Voters, in their zealotry against the perceived status quo, failed to realize how much of what we have right now you don’t want to cut recklessly, as well as jus
138.
by willy_k 1y ago
Shoutout Mullvad VPN, honorable mention to Tailscale (they had an acquire users phase and VC funding but a rug pull does not seem likely for the time being).
139.
by willy_k 1y ago
> it used to be so easy to get multiple experts on anything to answer your questions. It’s still possible to get useful answers on niche topics, but you will also get flooded with questionable sub-specific dogma, and god forbid your ques
140.
by willy_k 2y ago
You’re describing Twitter not conspiracy communities in general. On the UFO front at least I am aware of multiple YouTube channels and Discord servers with healthy diversity of thought, and I’m sure the same goes for other areas.
141.
by willy_k 2y ago
Something that has been making the sponsorship rounds now is Ground News[0] which I have found very useful with just the free tier. But given how many people I have seen sponsored by them, I wonder if there is some catch, especially because
142.
by willy_k 2y ago
Sigh, another libtard completely divorced from reality, suckered by the fake news. Those are reasons why everyone except MAGA is dumb, duh. /s
143.
by willy_k 2y ago
I believe they are talking about the SAVE act, which would apparently introduce new requirements for voter registration, supposedly making it so that identification and legal birth names would need to match in order to register to vote whic
144.
by willy_k 2y ago
Brains are already hackable in multiple senses. Namely through exogenous chemicals, electrical stimulation, hypnosis, and combinations of these. They aren’t necessarily reverse engineerable, which is what computational models like LLM-tied
145.
by willy_k 2y ago
We can simulate the three body problem because all of the significant parts of the problem can be known. This is FAR from the case when it comes to the brain.
146.
by willy_k 2y ago
> This is moving goalposts as it has little to do with my original premise. Your original comment was (replying to one) about “decompilation” of binaries. You seem to be talking about some fantasy perfect decompilation that gets variable
147.
by willy_k 2y ago
Is that any different from the bulk of standup today?
148.
by willy_k 2y ago
Wonder if there’s some pro-OpenAI system prompt getting in the way of that.
149.
by willy_k 2y ago
So they made Claude that knows a bit more.
150.
by willy_k 2y ago
That’s assuming that they would be subject to atmosphere and inertia, which contradicts observed UAP phenomena. (queue internet aneurysm?)

More ›