Considering the author of the text was an LLM (100% AI generated, not assisted on Pangram), I don't think so.
> Qt everywhere maybe? As someone who picked the KDE side decades ago, I have to say that would be for the best.
I was expecting proper green threads (it's not like it's impossible in C, there's several C libraries for doing it).
> Fable was used in several iterations to test for security issues before each release. How? The point of Fable (vs. Mythos) is that it has extreme guardrails against doing e. g. security work, even mentioning…
The most esoteric I've seen from an LLM was this light rendering fix in a TempleOS game engine: https://gitgud.io/CrunkLord420/cyberchud/-/commit/c49879751c... Though it's still cheating with a compatibility layer, I…
Did you actually manually review the 600k lines of code? And the reason LLMs tend to produce such insanely large codebases is partially what the comment you're replying to explained, they duplicate everything all over…
Considering wikimedia hosts what most people would consider porn (randomly came across a video of a couple going at it once) without any ability to filter it from the rest of the content/hide by default, it's not that…
Why use relatively ancient models like whisper and not e. g. parakeet?
There's tons, Parakeet was the last I remember seeing which seemed to gain traction (independent lightweight implementations etc).
If it was faster but worse, maybe compare it to a smaller whisper model?
Yes, this is something almost all other anti-bot/fraud prevention solutions already does, and there are already bypasses developed simulating human mouse movement. I expect this will be effective for maybe a day.
there are already stuff to trick mouse movement profiling, since its used by other bot protection stuff.
Didn't recaptcha v3 also do this? iirc the "silently monitor in the background" was part of the selling point.
You don't need to actually solve the PoW to bypass anubis, look up the source for the various "bypass anubis" browser extensions.
There's a ton of (known and reported, though google has made that difficult now) malicious extensions on the chrome web store, see e. g. https://palant.info/2025/01/13/chrome-web-store-is-a-mess/
I take it you either have a godly IP reputation, or you don't hit 90% of the current internet which is behind cloudflare, google, akamai, etc? They're pretty brutal in blocking non-browsers, I tend to have to add a lot…
didn't the capnproto author end up creating protobuf2? or am I mixing things up.
rust code tends to be incredibly slow unless you spend a ton of time optimizing (compared to other compiled languages, and contrary to popular belief).
Didn't the testifying UK academics explicitly say the current research was inconclusive?
Google's transformer paper was largely just stripping out parts of an earlier (now apparently largely forgotten) paper/model and running it on Google scale compute and data (which the original researchers obviously…
On the humans (vs. pangram) ability to reliably spot LLM generated content, there was an independent study: https://arxiv.org/html/2501.15654v1
> It was never science. There were credentialed academics running studies and contributing to the UI/UX design in KDE in the 2000s at least, though I guess your comment is true in the sense that (most) commercial…
it's just Yet Another Chromium, like everything else.
You can also just add -fwrapv.
in C/C++ you have __builtin_add_overflow(), __builtin_sub_overflow(), __builtin_mul_overflow()
Considering the author of the text was an LLM (100% AI generated, not assisted on Pangram), I don't think so.
> Qt everywhere maybe? As someone who picked the KDE side decades ago, I have to say that would be for the best.
I was expecting proper green threads (it's not like it's impossible in C, there's several C libraries for doing it).
> Fable was used in several iterations to test for security issues before each release. How? The point of Fable (vs. Mythos) is that it has extreme guardrails against doing e. g. security work, even mentioning…
The most esoteric I've seen from an LLM was this light rendering fix in a TempleOS game engine: https://gitgud.io/CrunkLord420/cyberchud/-/commit/c49879751c... Though it's still cheating with a compatibility layer, I…
Did you actually manually review the 600k lines of code? And the reason LLMs tend to produce such insanely large codebases is partially what the comment you're replying to explained, they duplicate everything all over…
Considering wikimedia hosts what most people would consider porn (randomly came across a video of a couple going at it once) without any ability to filter it from the rest of the content/hide by default, it's not that…
Why use relatively ancient models like whisper and not e. g. parakeet?
There's tons, Parakeet was the last I remember seeing which seemed to gain traction (independent lightweight implementations etc).
If it was faster but worse, maybe compare it to a smaller whisper model?
Yes, this is something almost all other anti-bot/fraud prevention solutions already does, and there are already bypasses developed simulating human mouse movement. I expect this will be effective for maybe a day.
there are already stuff to trick mouse movement profiling, since its used by other bot protection stuff.
Didn't recaptcha v3 also do this? iirc the "silently monitor in the background" was part of the selling point.
You don't need to actually solve the PoW to bypass anubis, look up the source for the various "bypass anubis" browser extensions.
There's a ton of (known and reported, though google has made that difficult now) malicious extensions on the chrome web store, see e. g. https://palant.info/2025/01/13/chrome-web-store-is-a-mess/
I take it you either have a godly IP reputation, or you don't hit 90% of the current internet which is behind cloudflare, google, akamai, etc? They're pretty brutal in blocking non-browsers, I tend to have to add a lot…
didn't the capnproto author end up creating protobuf2? or am I mixing things up.
rust code tends to be incredibly slow unless you spend a ton of time optimizing (compared to other compiled languages, and contrary to popular belief).
Didn't the testifying UK academics explicitly say the current research was inconclusive?
Google's transformer paper was largely just stripping out parts of an earlier (now apparently largely forgotten) paper/model and running it on Google scale compute and data (which the original researchers obviously…
On the humans (vs. pangram) ability to reliably spot LLM generated content, there was an independent study: https://arxiv.org/html/2501.15654v1
> It was never science. There were credentialed academics running studies and contributing to the UI/UX design in KDE in the 2000s at least, though I guess your comment is true in the sense that (most) commercial…
it's just Yet Another Chromium, like everything else.
You can also just add -fwrapv.
in C/C++ you have __builtin_add_overflow(), __builtin_sub_overflow(), __builtin_mul_overflow()