The e-zines can make a comeback, there's probably enough content now. phreak2030s: "put the local load in the bearing firmwares". Or something :)
Sorry, I thought you were asking about the tech applications. I have no idea about the other stuff, can't comment.
> what even are these experiences? whom are they for? One possible use for these is to facilitate lab -> this -> real world for robot models. Basically better than lab / algorithmically designed environments and cheaper…
IIUC, the first setup couldn't be exploited because the site didn't run any php. You kinda need the server to run some php if you're gonna use a php vuln. It got solved ~2h after they added a php-version.php ...
> (so you might not ever see a direct refusal) A note that the only inference providers that hide thinking are the closed labs. Most (all, afaik) inference providers serving open models will send the clear thinking…
Sorry, I'm not following. Constrained generation is something that helps you, the caller, to get back a message that looks a particular way. I'm having trouble seeing how this would work in an adversarial way. Perhaps…
> but massaging the sampler(?) also sounds like that might give a lot of control over outputs with the weights staying technically the same? Yes, this is what constrained decoding means. You constrain the model to only…
> So they admit that they could do more, but they won't? Yeah, there's another paragraph: > Identifying and removing cyber-related data. During development, we observed vulnerability research performance improving…
I think the golden "catch 22" error will forever be the classic "No keyboard detected, press F1 to continue".
> so you say they only sat on it for a week? This is absurd, and it shows that you have absolutely 0 idea of how any of this works. Finding a bug and delivering a working RCE 11 days later is absolutely not "sitting on…
No idea what your comment is even about. The researchers that found and reported this did not sit on this for 2 months. > 1 August 2026 [...] Ryan Austin finds the RCE paths, tweets a cryptographic hash on 5 August, and…
> This is just telling the chatbot to lie. Why do you think that? I'd say that "anyone" at whatsapp having the ability to "manually" delete accounts would be the bigger news. In large corps, no one should have that…
> The biggest LLM sellouts in open source I'm pretty sure someone will soon call Linus a sell-out. The hate / denial / radicalisation is just sad at this point... Do, or don't, but at least don't hate the others. Be and…
> I am just not sure what it measures. All of the above, to the extent that they can be measured from a simple to verify end condition (stats, win rates, etc). Adaptability, ability to understand and progress on a task,…
Heh, true. My brain went into "pre space" and "post space" as halves :)
Anthropic doing free advertising for glm and other open models :D Impressive capabilities shown on benchmarks don't really make a difference if you can't use them. I'll take a lower chance of success over constant…
> Can a model be trained purely on LLM-Outputs and access to tools such as a Turing complete environment? Yes, it can. The Phi series (1 and 2 specifically) were trained by MS as a test of this. Both pretrain and…
Yeah, I should have worded that better. I'd say it's at the upper level of what can be ran on plain consumer hw (like a laptop, mac, etc). For ~k3 you'd realistically need dedicated hw anyways. Or a cluster of consumer…
> as a clear sign they are scared shitless of local models Partially agree and partially disagree. Qwen 3.8 next (120b + 50b ngrams) is at the upper end of a local model, and serving it at good speeds (esp for pp) is…
> I'm really challenged to see the benefit here. > write a more generalized SDR application for this hardware Yeahhh, no. Sorry but the entry barrier is now extremely lowered, especially in effort. ~20 years ago this…
> RSI fad dies down Don't fall for the "marketing hype" antihype. RSI is a trendy term today, and lots of people discuss it, but it's nothing new. We've basically had forms of RSI since the start of the LLM craze.…
Part of it is probably because we've already been through this? The SWE field has seen exactly these reactions over the past 4+ years. Some people saw "the spark" before LLMs were mainstream. We had code autocomplete…
> AI is kinda motivating everyone to do everything on their own instead of gathering up and contributing to one (or two) projects, which is something I'm worried about in Open Source software as a whole. A huge reason…
I think you hallucinated the superconductor part. Which is funny in context, but understandable because that's what I also thought while reading the title. I had to do a double-take on the word after room-temperature,…
> yearly meetup for them in Lindau IIRC Oppenheimer facilitated / sponsored a meeting there after the Manhattan Project, as a gathering where physicists could talk freely about physics for a change.
The e-zines can make a comeback, there's probably enough content now. phreak2030s: "put the local load in the bearing firmwares". Or something :)
Sorry, I thought you were asking about the tech applications. I have no idea about the other stuff, can't comment.
> what even are these experiences? whom are they for? One possible use for these is to facilitate lab -> this -> real world for robot models. Basically better than lab / algorithmically designed environments and cheaper…
IIUC, the first setup couldn't be exploited because the site didn't run any php. You kinda need the server to run some php if you're gonna use a php vuln. It got solved ~2h after they added a php-version.php ...
> (so you might not ever see a direct refusal) A note that the only inference providers that hide thinking are the closed labs. Most (all, afaik) inference providers serving open models will send the clear thinking…
Sorry, I'm not following. Constrained generation is something that helps you, the caller, to get back a message that looks a particular way. I'm having trouble seeing how this would work in an adversarial way. Perhaps…
> but massaging the sampler(?) also sounds like that might give a lot of control over outputs with the weights staying technically the same? Yes, this is what constrained decoding means. You constrain the model to only…
> So they admit that they could do more, but they won't? Yeah, there's another paragraph: > Identifying and removing cyber-related data. During development, we observed vulnerability research performance improving…
I think the golden "catch 22" error will forever be the classic "No keyboard detected, press F1 to continue".
> so you say they only sat on it for a week? This is absurd, and it shows that you have absolutely 0 idea of how any of this works. Finding a bug and delivering a working RCE 11 days later is absolutely not "sitting on…
No idea what your comment is even about. The researchers that found and reported this did not sit on this for 2 months. > 1 August 2026 [...] Ryan Austin finds the RCE paths, tweets a cryptographic hash on 5 August, and…
> This is just telling the chatbot to lie. Why do you think that? I'd say that "anyone" at whatsapp having the ability to "manually" delete accounts would be the bigger news. In large corps, no one should have that…
> The biggest LLM sellouts in open source I'm pretty sure someone will soon call Linus a sell-out. The hate / denial / radicalisation is just sad at this point... Do, or don't, but at least don't hate the others. Be and…
> I am just not sure what it measures. All of the above, to the extent that they can be measured from a simple to verify end condition (stats, win rates, etc). Adaptability, ability to understand and progress on a task,…
Heh, true. My brain went into "pre space" and "post space" as halves :)
Anthropic doing free advertising for glm and other open models :D Impressive capabilities shown on benchmarks don't really make a difference if you can't use them. I'll take a lower chance of success over constant…
> Can a model be trained purely on LLM-Outputs and access to tools such as a Turing complete environment? Yes, it can. The Phi series (1 and 2 specifically) were trained by MS as a test of this. Both pretrain and…
Yeah, I should have worded that better. I'd say it's at the upper level of what can be ran on plain consumer hw (like a laptop, mac, etc). For ~k3 you'd realistically need dedicated hw anyways. Or a cluster of consumer…
> as a clear sign they are scared shitless of local models Partially agree and partially disagree. Qwen 3.8 next (120b + 50b ngrams) is at the upper end of a local model, and serving it at good speeds (esp for pp) is…
> I'm really challenged to see the benefit here. > write a more generalized SDR application for this hardware Yeahhh, no. Sorry but the entry barrier is now extremely lowered, especially in effort. ~20 years ago this…
> RSI fad dies down Don't fall for the "marketing hype" antihype. RSI is a trendy term today, and lots of people discuss it, but it's nothing new. We've basically had forms of RSI since the start of the LLM craze.…
Part of it is probably because we've already been through this? The SWE field has seen exactly these reactions over the past 4+ years. Some people saw "the spark" before LLMs were mainstream. We had code autocomplete…
> AI is kinda motivating everyone to do everything on their own instead of gathering up and contributing to one (or two) projects, which is something I'm worried about in Open Source software as a whole. A huge reason…
I think you hallucinated the superconductor part. Which is funny in context, but understandable because that's what I also thought while reading the title. I had to do a double-take on the word after room-temperature,…
> yearly meetup for them in Lindau IIRC Oppenheimer facilitated / sponsored a meeting there after the Manhattan Project, as a gathering where physicists could talk freely about physics for a change.