HNHacker News
TopNewBestAskShowJobs

popalchemist

611 karma · joined July 17, 2022

submissionscomments
popalchemist··on Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse
Daniel, he openly and proudly flaunts the possibility that what he is doing will possibly destroy the human race. Verbally acknowledging that is not cheap rhetoric.

Let me ask you this - to put a fine point on your policy - under what circumstances would discussion of psychotic behaviors (not pretenses to a person's diagnosis but discussion of behaviors as psychotic or possibly indicative of psychosis or other forms of insanity) be permitted?

Because if your rule is a blanket rule that all such discussion is illegal here, then you have created a safe haven where the most destructive personalities in the world are the most protected here. And they just happen to be people your employer is directly invested in and bound to. Hm.

Do you see the problem?

I think it's pretty obvious that the policies may be constructed to appear neutral, and perhaps they were conceived that way, but structurally and materially what they do is prevent and downright squash legitimate dissent, evaluation, and critique of horrible things happening in this world, of which YC is a part. I would very much like to see that truth acknowledged or the policy amended to prevent the deliberate silencing of fair critique of people like Altman.

If your attitude toward this issue is that the language policing supercedes this concern, then that is tantamount to an admission of being part of the oligarchy's apparatus and, in whatever small way, hacker news has become an arm of the fascist regime. (I suspect this is the case but I invite a direct response.)

popalchemist··on Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse
This is the only rational stance on this.
popalchemist··on Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse
The man is clearly NOT NORMAL. Seeing as he seemingly admits that his own actions at OpenAI might end the human race every few days.

Flippant about self-annihilation / extinction. What possible indicator could be more explicit that someone is NOT NORMAL?

https://www.newyorker.com/magazine/2026/04/13/sam-altman-may...

combine that with the fact that nearly every single person interviewed for the NYT profile on him said he is a liar...

For me, that is enough reason to take the allegations seriously. They are commensurate with what is plainly observable about him.

popalchemist··on Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse
What I said is not a diagnosis and no intelligent person would think so. The word psychotic is not ONLY a diagnostic term, it has a colloquial meaning, which aptly covers the fact that he is seemingly happy to destroy the world's economy and possibly even kill us all. But you knew that, dang. If you know a better word that more accurately speaks to that exact type of insanity that he is obviously displaying, I am open to feedback.

Perhaps just EVIL? Very curious to hear what you come up with.

By the way it is interesting how often hacker news's editorial staff uses language policing to hide commentary against a specific set of people who have been involved in YC. Not lost on anyone here, certainly not me.

popalchemist··on Show HN: jpm – a JavaScript package manager in Rust, every line by Claude Code
the reason the authors of pnpm know something works as intended is because every line of code came from them and they can hold the model of its mechanics in mind. you can't.

I'm not particularly concerned about this - I am never going to use your thing.

But your answer is insufficient.

popalchemist··on Sam Altman's sister amends lawsuit accusing OpenAI CEO of sexual abuse
[flagged]
popalchemist··on Show HN: jpm – a JavaScript package manager in Rust, every line by Claude Code
how do you know it works as intended if you didn't write it

the delivery of code is a mission critical feature; if there is a security hole you could be legally liable, you know that right?

popalchemist··on We're going to need default hard budget caps on pretty much everything
Utilities required for basic survival are expected to remain on for obvious reasons. Like not letting people die.
popalchemist··on Agents don't need memory, they need documentation
There is a hierarchical distillation of understanding in human processes that agentic processes do not yet execute.
popalchemist··on FLUX 3 Image
true but they're announced https://x.com/ideogram_ai/status/2105327223431737780
popalchemist··on FLUX 3 Image
https://arxiv.org/html/2603.28489v1

Video Generation Models as World Models: Efficient Paradigms, Architectures and Algorithms

https://bfl.ai/models/flux-3-action

The idea to use predictive models of future observations for decision making has a long history, including early demonstrations of robot control which combined action-conditioned video prediction with model-predictive control.

Clearly I'm the uneducated one here eh

popalchemist··on Show HN: Social SDK, a typed TypeScript toolkit for 7 social platforms
was googling for exactly something like this yesterday. thank the maker!
popalchemist··on Every SaaS business will become a harness around a model
Is the article about UX? I think it's talking about DX. The internals of the company/software.
popalchemist··on FLUX 3 Image
They are proto world models (lots written about this - flux being an example of a video model whose weights also power world-action-engines used in robots) in that they attempt to model causality in time, the thing that is required for what OP is asking for and which image models will never do because it is out of domain.
popalchemist··on FLUX 3 Image
Image edit models can probably do a grid, but the temporal accuracy / coherence will never match what a video model, which is really a world model, can do.
popalchemist··on How accurately calibrated is Jev?
not very
popalchemist··on FLUX 3 Image
The task you're describing is a video model task, not an image model task. It's inherently temporal.

Generate a sprite in an image editor, then use a video model to make the loop you want; then turn the resulting video back into individual sprite images.

popalchemist··on FLUX 3 Image
Ideogram 4.5 released 2 days ago with more features along these lines

https://www.youtube.com/watch?v=2mecWZgbaEg

popalchemist··on Supabase is acquiring Turso
I feel like this makes a lot of sense and is a rare win-win-win (supabase-turso-customers being the parties).
popalchemist··on The death of web development education
What does that mean in this context?
popalchemist··on The death of web development education
The author posits a situation where big-tech AI firms teaming up with fascists as a hypothetical.

It is already here. That is the current reality.

popalchemist··on Pi 1.0
Does this replace Claude Code in VS Code entirely?
popalchemist··on Returning from vacation? The government can search your phone without a warrant
Lockdown mode prevents access via USB as well.
popalchemist··on Responsible Release of AI-Generated Mathematics
It directly undermines human achievement - something some people have spent a lifetime in pursuit of - for the sake of marketing. OpenAI gains nothing at all; the mathemeticians on the verge of achieving a lifelong pursuit have their entire career trajectory and perhaps reason for being eviscerated without a thought. If the goal is to further mathematics, the mathemeticians are the ones who would best understand whether what OpenAI is doing is helping or hurting.
popalchemist··on It's Time to Investigate the AI Labs
The r-word policy is noted. The effect it has had is a cooling effect - I am less likely to contribute to this community now than I was before; this is how the site leans more and more right-wing every day, by the way.

What matters to me is people who are robotically enacting policies that are anti-human in the name of capital come back to their humanity. That's why I took the time with you.

popalchemist··on It's Time to Investigate the AI Labs
I am not a jerk for using language you don't like. For me, it is a civic calling.

You would know that if you read the post I thoughtfully constructed to enlighten you on that point.

Legitimate conflicts deserve to be met with legitimate resolutions. Your ban-hammer attitude is the thing out of line here. That's why I keep calling you authoritarian. You are acting like the sole arbiter of what is good and acceptable and the rigidity with which you do it has a cooling effect on real, legitimate discourse. Which is anti-human of you.

popalchemist··on It's Time to Investigate the AI Labs
Then why not just edit the one word and give a warning shot instead of censoring the rest of the post, which was entirely on topic and argued well, and in good faith?
popalchemist··on It's Time to Investigate the AI Labs
It's not deflecting. You're moralizing. You're not the thought police, Tom. How did you get to this point?

But you do you, I already knew this place is an exercise in authoritarianism, you're just proving it.

popalchemist··on It's Time to Investigate the AI Labs
[flagged]
popalchemist··on It's Time to Investigate the AI Labs
Is it literally the one word?

Because you could just change it to "stupid" or "asinine" and the comment remains completely on-topic. Feels rather like censorship, I think you're being disingenuous.

← PreviousPage 2 of 22Next →