sdwr

7 ms·

31.
by sdwr 7mo ago
The more they share, the easier it is to exploit the system.
32.
by sdwr 7mo ago
Great question! Clean separation between Gemini Pro and the other answers
33.
by sdwr 7mo ago
Baseline demand affects the numbers much more than the unpredictable spikes do. You can come up with edge cases if you like, but the reality is that it all averages out pretty well with large volume.
34.
by sdwr 7mo ago
And how are you supposed to guarantee equivalent functionality by analyzing "README files, API docs, and type definitions"?
35.
by sdwr 7mo ago
It's very rational to overshoot in that situation. If you build your lifestyle and then FIRE, you are derisking your budget while you still have income. But wanting to change your lifestyle when you retire is incredibly risky, especi
36.
by sdwr 7mo ago
Taste is definitely not overindexing on "it's not x, it's y" and accusing everyone who makes a comparison of being a bot
37.
by sdwr 7mo ago
It's like any good cryptographic challenge - easier to validate the answer than to come up with it from scratch
38.
by sdwr 7mo ago
Yeah, the execution is stronger than the idea and design choices as well, which suggests heavy AI support
39.
by sdwr 7mo ago
It's a formalized, sanitized simulacrum of striving! Like sports is for competition
40.
by sdwr 7mo ago
That honestly sounds great! If it works...
41.
by sdwr 7mo ago
If you know your exact use case, have prior work to build on, think deeply and extensively about the problem domain, and don't need competitive results, you can save a lot of lines of code!
42.
by sdwr 7mo ago
It's not bi-directional. When you friend jacques, you see people who he has friended as FoaF, not people who have friended him. It's a curator-style system, not a direct popularity one.
43.
by sdwr 8mo ago
It's a fine argument, just rough around the edges: Human brains use redundancy and the physical independence of neurons to build new pathways over time. Current LLMs have no redundancy and brittle weights. Their technology and architec
44.
by sdwr 8mo ago
The perfect information version calculates the path once on click, the local version needs to recalculate every time the unit moves (and new tiles are "seen")
45.
by sdwr 8mo ago
Maybe human part inspectors should get treats for inspecting parts correctly
46.
by sdwr 8mo ago
Awesome idea! The entity tracking is very exciting, most interesting part imo I think the budget is noticeable in the sentiment analysis unfortunately, the tags and entity recognition are good but the sentiment ratings themselves seem prett
47.
by sdwr 8mo ago
Yeah, or just the steam text guides would be a huge advantage. I really doubt it's playing completely blind
48.
by sdwr 8mo ago
Gwern is the nerd's nerd
49.
by sdwr 8mo ago
I'm sure the goal is that reading files is something you debug, not monitor, like individual network requests in a browser.
50.
by sdwr 8mo ago
An error message helps people skirt the restriction, by providing immediate feedback on what does/doesn't get flagged. Same idea as shadow banning, ban waves, and generic errors for sensitive actions
51.
by sdwr 8mo ago
> arXiv "emergent" (the count of AI papers about emergence) has a clear, unambiguous R² maximum. The other four are monotonically better fit by a line The only metric going infinite is the one that measures hype
52.
by sdwr 8mo ago
What does clean room mean in this context? They built it from the assets with the game as a reference , but didn't look at the engine source code?
53.
by sdwr 8mo ago
It's very indirect. The message is "the government is soft on fraud, partially because of liberal values", but the author does everything possible to not actually say it.
54.
by sdwr 8mo ago
All I remember hearing about this is how creepy and racist it was to look for kids at a black-owned daycare. It was a scam the whole time?
55.
by sdwr 8mo ago
An AI "friend bot" could talk like an individual, but be controlled by a network that is making global decisions - what if one of its goals is reducing social unrest by undermining the confidence of users with certain political vi
56.
by sdwr 8mo ago
I don't think it's pedantic, he's trying to make a broad point about his mentality, using the detail as the defining example
57.
by sdwr 8mo ago
Informal, casual, friendly
58.
by sdwr 8mo ago
It's absolutely the correct distinction to draw. AI will be able to present as a person while behaving as a system. We don't have intuitions for dealing with that.
59.
by sdwr 8mo ago
They are way better at code-related tasks than design or strategy ones. Anything involving users or business strategy or a "why" is vague and misguided, they have no insight.
60.
by sdwr 8mo ago
I think this is a backwards approach, especially for children. Gen AI is magical, it makes stuff appear out of thin air! And it's limited, everything it makes kinda looks the same And it's forgetful, it doesn't remember what

More ›