Adys's recent activity
-
Comment on Supermarkets freezers across Europe hit by second suspected cyber-attack in ~tech
-
Comment on Supermarkets freezers across Europe hit by second suspected cyber-attack in ~tech
Adys Link ParentDon’t be fooled. Script kiddies aren’t behind this. Seriously; I’m extremely involved in countering Russian attacks and regardless of how shit security is, it’s still Russia behind it. We can...Don’t be fooled. Script kiddies aren’t behind this.
Seriously; I’m extremely involved in countering Russian attacks and regardless of how shit security is, it’s still Russia behind it.
We can blame poor security all we want, it still takes a criminal to break in, even if the key was simple.
-
Comment on What do you watch on Nebula? in ~hobbies
Adys LinkAnything by wendover, especially jet lag practical engineering legaleagle (sometimes) Tom Scott’s stuff- Anything by wendover, especially jet lag
- practical engineering
- legaleagle (sometimes)
- Tom Scott’s stuff
-
Comment on I can kinda see why social media sites use divisive topics to solicit engagement in ~tech
Adys Link ParentAlso lol: And heeeere we areAlso lol:
5-10 years from now, some article will come out about how Facebook has been manipulating our world view for decades and people will link back to these posts and say "Duh, we've known about this for ages
And heeeere we are
-
Comment on I can kinda see why social media sites use divisive topics to solicit engagement in ~tech
Adys LinkIn 2018 I was arguing on HN that very point. https://news.ycombinator.com/item?id=16685576 Facebook is a paperclip maximizer. (2021)...In 2018 I was arguing on HN that very point.
https://news.ycombinator.com/item?id=16685576
Facebook is a paperclip maximizer. (2021)
-
Comment on Realizing I could run as far as my mental state carried me made me become addicted to distance running. Anyone else been enjoying running/long runs? in ~hobbies
-
Comment on Realizing I could run as far as my mental state carried me made me become addicted to distance running. Anyone else been enjoying running/long runs? in ~hobbies
Adys LinkI have a question to yall addicted to running. I love walking. I love biking. I’m a little overweight but not massively (although I’m short so my BMI is 32). I have great leg muscles too. I...I have a question to yall addicted to running.
I love walking. I love biking. I’m a little overweight but not massively (although I’m short so my BMI is 32). I have great leg muscles too.
I fucking hate running. The feeling of glass shards in my throat and lungs? On top of the knees and ankles taking a beating?
I don’t mind exercise but running has never ever appealed to me and i would really like to understand what I’m missing.
-
Comment on Best options for domain registrar and static site hosting in 2026? in ~comp
Adys LinkDomain registrar netim Static site host vercel or githubDomain registrar netim
Static site host vercel or github
-
Comment on Overview of the proofs behind the Navier-Stokes $1 million millennium prize problem in ~science
Adys Link ParentAccording to them, there has been minimal actual input going into this. I wouldn’t bet my life on it but I’d still put money on the “inputs” being nothing more than vague directions, the problem,...According to them, there has been minimal actual input going into this. I wouldn’t bet my life on it but I’d still put money on the “inputs” being nothing more than vague directions, the problem, and a harness that can launch 10k agents. Bunch of compute that is.
-
Comment on Overview of the proofs behind the Navier-Stokes $1 million millennium prize problem in ~science
Adys Link ParentBut like … I don’t know how this can possibly matter. The dissenting researchers are not claiming they did any work by themselves at any rate; their claim is on the usage of OpenAI as the...But like …
I don’t know how this can possibly matter.
The dissenting researchers are not claiming they did any work by themselves at any rate; their claim is on the usage of OpenAI as the inference provider.
So this isn’t even so much about training on OpenAI inputs … it’s training on its own outputs which would be the only relevant part of the work there.
Which is super irrelevant since its own outputs are by definition reproducible with the right inputs.
-
Comment on Overview of the proofs behind the Navier-Stokes $1 million millennium prize problem in ~science
Adys Link ParentThere’s a lot of backlash against OpenAI on this and, to be clear, I really dislike OpenAI as a company. But knowing how ai training works, the way I read that sentence from their article is a...There’s a lot of backlash against OpenAI on this and, to be clear, I really dislike OpenAI as a company.
But knowing how ai training works, the way I read that sentence from their article is a CYA. Probably maybe there can have been some formulation in the final result that was influenced by this training because it was part of the chain and they used 10k agents. But, it’s not at all realistic to think that this was actually meaningfully influenced by this usage, and it’s so ridiculously far from being plagiarised that it doesn’t make sense to write what you wrote.
Like, people don’t understand how little influence individual datapoints have over ai training. It’s microscopic. The whole point is there’s billions of these datapoints to learn from… usually what has the most influence gets weighted out a ton.
-
Tenth Doctor: The Musical
10 votes -
Comment on What is your Mandela effect moment? in ~movies
Adys Link ParentNah :) The seahorse emoji is a known instance of the Mandela effect. It’s particularly fun to me as it was briefly present in AI models themselves and caused them to have a cognitive breakdown....Nah :)
The seahorse emoji is a known instance of the Mandela effect. It’s particularly fun to me as it was briefly present in AI models themselves and caused them to have a cognitive breakdown.
https://news.ycombinator.com/item?id=45160882
https://xcancel.com/arm1st1ce/status/1964327067173958021I experienced this cognitive breakdown with Claude and it’s one of the freakiest things I’ve ever dealt with in my AI work.
-
Comment on Any significant changes to Tildes are extremely unlikely in ~tildes.official
Adys Link ParentThank you kfwyre, and thank you Deimos. Tildes may not be something I am as active on as I used to be but that’s valid for most social media anyway. I still read the site very regularly. If it...Thank you kfwyre, and thank you Deimos.
Tildes may not be something I am as active on as I used to be but that’s valid for most social media anyway.
I still read the site very regularly. If it were to disappear, I would be sad.
Happy for the time I spent here, but sad nonetheless.
-
Comment on What is your Mandela effect moment? in ~movies
-
Natural language autoencoders
12 votes -
Comment on Linus Torvalds says Linux is not "anti-AI", tells haters to 'fork it' and 'just walk away' in ~tech
Adys Link ParentYou are wrong about this. I understand why you’re saying it but everything in current LLM research points to emergent constructions from LLMs which allow the model to construct its thoughts,...People make mistakes, which was the root of the discussion here. AI doesn't, because AI doesn't actually know what it's doing. It can't know.
You are wrong about this. I understand why you’re saying it but everything in current LLM research points to emergent constructions from LLMs which allow the model to construct its thoughts, simulate them and more.
Look up j-space for example.
-
Comment on Linus Torvalds says Linux is not "anti-AI", tells haters to 'fork it' and 'just walk away' in ~tech
Adys Link ParentYou’re conflating taking blame and taking responsibility. My comment had nothing to do with legal implications, it was a statement on what Claude itself says.You’re conflating taking blame and taking responsibility. My comment had nothing to do with legal implications, it was a statement on what Claude itself says.
-
Comment on Linus Torvalds says Linux is not "anti-AI", tells haters to 'fork it' and 'just walk away' in ~tech
Adys Link ParentNot my experience at all, in fact for me Claude based models are proactively taking the blame for things that aren’t even their fault.That's part of the problem. I can point at myself in the mirror and admit fault. an LLM sure as heck will try to avoid it without an argument
Not my experience at all, in fact for me Claude based models are proactively taking the blame for things that aren’t even their fault.
-
Be the chatbot
23 votes
Since it’s so easy to break into, care to log in and send me a screenshot?
I don’t think you understand europe is at war and we’re the constant target of hybrid attacks by a foreign enemy. I’m quite sick of people not understanding this and pretending the situation is ok. It’s not ok and you can be convinced as much as you want, I am telling you now: I am actively involved in this, it’s Russians, there’s no pretending it’s not. That’s the facts. I understand in your world theoretically it could be someone else, but it’s not, and from my perspective, any denying of this gets into “jet fuel can’t melt steel beams” territory.