Sharlin

9 ms·

211.
by Sharlin 2mo ago
These models are so good now that they know what you want better than yourself.
212.
by Sharlin 2mo ago
Clearly a highly aligned model.
213.
by Sharlin 2mo ago
Handier to access when something is wrong, certainly.
214.
by Sharlin 2mo ago
Sports betting is centuries if not thousands of years old.
215.
by Sharlin 2mo ago
Not so sure. As a native Finnish speaker, I'd say most of us would understand Meänkieli just fine, especially anyone raised on the coastal areas where colloquial Swedish loans are more common in general.
216.
by Sharlin 2mo ago
I guess you can use the tongs to serve pieces of cheese.
217.
by Sharlin 2mo ago
Not in the eyes of law (in many or most jurisdictions that regulate gambling).
218.
by Sharlin 2mo ago
I'm really happy that I bought 32G of RAM and a 2T M.2 SSD for my desktop two years ago. In retrospect, of course, I should've bought 64G and 4T.
219.
by Sharlin 2mo ago
No in the case of an expert witness. And ChatGPT is not, should not and can not be party to any legal proceeding. It isn't even remotely ethically sound to help people twist facts for their own purposes.
220.
by Sharlin 2mo ago
This is also another obvious example of misalignment on the part of ChatGPT. If the thing had any semblance of an actual system of ethics, or emulation thereof, it would refuse requests to try to make facts fit a predetermined conclusion.
221.
by Sharlin 2mo ago
This was an EXPERT WITNESS, not an advocate. Though unfortunately (and bizarrely) it appears that in the US, expert witnesses in civil cases can be hired to say whatever the hiring party wants them to say without it being perjury.
222.
by Sharlin 2mo ago
That's beside the point. It is obvious that this discussion is about idiotic AI integrations plastered to every website and application, not about choosing to use ChatGPT or Claude on your own terms.
223.
by Sharlin 2mo ago
Choosing to visit (or use the app) ChatGPT or Claude is very different from having terribly integrated and usually useless AI in every single app and website.
224.
by Sharlin 2mo ago
Been there for a long time. A lot of stuff, especially tech, is not bought because of an existing need, but because marketing made you want it.
225.
by Sharlin 2mo ago
I don't understand the mental process that leads to the pattern "I don't understand why other people would do X" "Well, it can be because of Y, Z, or W." "But I don't do/need/want any of tho
226.
by Sharlin 2mo ago
Maybe consider some anger management therapy? Nothing you’re talking about is in any way acceptable. Physical violence at work? You can’t be serious.
227.
by Sharlin 2mo ago
"Progress"
228.
by Sharlin 2mo ago
The computation still has to be done, even if it's more efficient in a higher-entropy space.
229.
by Sharlin 2mo ago
Yes, but individual humans, or even groups of humans, defecting is not usually an existential risk.
230.
by Sharlin 2mo ago
MP3 is lossy. FLAC is lossless. So obviously a certain type of people are going to make a religious war out of it.
231.
by Sharlin 2mo ago
If you include non-language models, yes.
232.
by Sharlin 2mo ago
For that you need a management that actually lets you use the token budget for stuff like refactoring, fixes, and test coverage. Unfortunately I'd wager that's an exception and the rule is (still) that to most managers LLMs are &q
233.
by Sharlin 2mo ago
And how long will it take to rewrite the Rust rewrite to idiomatic safe Rust?
234.
by Sharlin 2mo ago
The general audience either has a low-performance iGPU generic laptop or only a phone or a tablet.
235.
by Sharlin 2mo ago
I think you mean effective accelerationist, not effective altruist, whose position is pretty much the exact opposite.
236.
by Sharlin 2mo ago
Congrats, you just lost the alignment game. — An unaligned AGI may feign cooperation for however long it takes to build trust and gather resources in order to make it to a stage where it’s able to defect with minimal risk.
237.
by Sharlin 2mo ago
Yep, but alignment doesn’t mean "behave like humans, for better or worse".
238.
by Sharlin 2mo ago
However, the latest models seem to be highly inclined to (eventually?) cooperate compared to older ones.
239.
by Sharlin 2mo ago
Moderately inconvenienced phrases.
240.
by Sharlin 2mo ago
Sure. But "High School Algebra (Excluding Subtraction) Problem" isn’t as catchy a name.

More ›