jamilton

7 ms·

61.
by jamilton 7mo ago
An obvious issue with the metaphor that comes to mind is that if you consider yourself to have a pretty good life, to be overall happy and satisfied, but you think it's possible to have an objectively much better life, then you'd
62.
by jamilton 7mo ago
I don’t think that assumption is being made, why do you think that? In terms of metaphor, training a model could be considered both knowledge acquired after birth and its evolution. But I don’t think it’s particularly useful to stay thinkin
63.
by jamilton 7mo ago
Technically any market that's about someone doing something by a certain time can be an assassination contract, if you think the market will it enforce it that way. Can't do it if they're dead.
64.
by jamilton 7mo ago
I would assume it was manually coded.
65.
by jamilton 7mo ago
They're all pretty common AI-isms.
66.
by jamilton 7mo ago
Still worth trying, IMO. An alternate approach that would be seen as consumer and business-friendly would be subsidizing companies with a certain level of fuel efficiency per passenger mile, targeted above current levels.
67.
by jamilton 7mo ago
Website looks nice. The copy being so painfully AI-written is a turnoff, enough that my first thought was "oh yeah, I remember hearing about this kind of app, I should look at the other one I'm thinking of". I do like that it
68.
by jamilton 7mo ago
Yes, it's a UX thing. You'd still have to edit it by typing afterwards as well. Similarly, raw LLM/chat interfaces are usually not the best option.
69.
by jamilton 8mo ago
People can tell. The premise is false. It’s sometimes hard to tell, obviously it’s hard to ascertain false negatives and false positives, but it’s usually pretty obvious.
70.
by jamilton 8mo ago
Feels very AI written in a way that makes it annoying to read with all the repetitive short sentences. Neat concept though, would be cool to see some tests of performance on some tasks.
71.
by jamilton 8mo ago
Cool. How’d you pick decks?
72.
by jamilton 8mo ago
Neat! I briefly tried digitizing some old VHS tapes for my family. I was just planning on giving them the files, maybe putting it on iCloud, it's a much smaller collection (and I don't have a NAS already!). I did a few, the time i
73.
by jamilton 8mo ago
The joke is presumably ancient vampires doing it in secret to be alive for millennia.
74.
by jamilton 8mo ago
>The public begins to associate blood transfusion with eccentric billionaires rather than with undead predators. This is a critical narrative shift. Not much of a shift...
75.
by jamilton 8mo ago
That quote was for the non-realtime model.
76.
by jamilton 8mo ago
It could have hands that feel but no vision, I think they were getting at that they thought embodiment and playing games in the modality of humans, without thousands of hours of play to reach competency, would be an important milestone.
77.
by jamilton 8mo ago
(Light spoilers) Supposedly the design intent was that you're not supposed to stealth in the dark much at all! It's really not obvious though, so it's a bit of a failure of design in that particular aspect. Interview with a d
78.
by jamilton 9mo ago
Yeah, convert to embedding, check if it's within a certain distance to an existing embedding and if so store it with that cluster and increment? Then check check further entries against against an average so clusters don't increas
79.
by jamilton 9mo ago
Being able to voice clone with PocketTTS seems major, it doesn't look like there's any support for that with Kokoro.
80.
by jamilton 9mo ago
Because people would just feel bad about it and keep doing it. I don't care about people feeling bad about immoral behavior, I care about them not doing it in the first place.
81.
by jamilton 9mo ago
https://openai.com/index/strengthening-chatgpt-responses-in-... "..our initial analysis estimates that around 0.15% of users active in a given week have conversations that include explicit indicators of potential
82.
by jamilton 9mo ago
Why that ratio in particular? I wonder if there’s a more complex ratio that could be better.
83.
by jamilton 10mo ago
Lol - they really should be locking down their email accounts and enforcing that policy. Or manually reviewing outbound messages before they can be sent. It seems likely that just telling the LLMs that will have a non-zero failure rate.
84.
by jamilton 10mo ago
The linked GitHub readme says it outputs a powerpoint file of the layers.
85.
by jamilton 10mo ago
What do you like about Void? It reads about how I would expect a base chat model to post.
86.
by jamilton 10mo ago
Yeah, I think it's an uncommon word. It's not a concept that would come up for most American English speakers, unless you're in a community that uses a language with another writing system (I think I first encountered it in a
87.
by jamilton 11mo ago
If AI is at the point where it is exactly as capable of your average junior 3D professional in 10 years, it will probably have automated a ton (double digit percentage?) of current jobs such that nothing is safe. There's a lot of compl
88.
by jamilton 11mo ago
Yeah, LLMs used to not be up to par for new Project Euler problems, but GPT-5 was able to do a few of the recent ones which I tried a few weeks ago.
89.
by jamilton 11mo ago
Yeah. I wonder why that is - is it because it highlights a conflict between our actions and values? If left unexamined, it's a non-issue, so having it spelled out feels like a problem being created?
90.
by jamilton 11mo ago
I know this has been said many times before, but I wonder why this is such a common outcome. Maybe from negative outcomes being underrepresented in the training data? Maybe that plus being something slightly niche and complex? The screensho

More ›