Hacker News: throw310822

New comment by throw310822 in "Anthropic CEO says the way for AI to win over the public is to cure cancer"

throw310822 — Mon, 17 Aug 2026 01:29:37 +0000

He's saying "people had enough of us saying AI will cure cancer, what we need to do to regain their trust is to deliver on our promises".

HN reaction: he makes absurd promises, he's disconnected from reality.

New comment by throw310822 in "What happens when an LLM never sees material beyond fifth grade?"

throw310822 — Sun, 16 Aug 2026 08:11:39 +0000

[February 2020]

New comment by throw310822 in "There's a reason why Israel has done little to rein in settler violence"

throw310822 — Sat, 15 Aug 2026 21:18:13 +0000

Of course. It's all so simple: Israel is asked to police itself against the misappropriation of highly desirable territories belonging to others. What do you think will happen- and not just over one year, but over decades? Individuals are not of one mind enough to be trusted in such a situation. So imagine states, that are composed not of one mind but of thousands, millions of individual wills.

The fundamental deception is the idea that it could ever work.

New comment by throw310822 in "Qwen 3.8 27B"

throw310822 — Fri, 14 Aug 2026 20:24:10 +0000

It's the singularity, baby.

New comment by throw310822 in "Trentino DataMine Opens 6MW Data Center Inside Active Dolomites Mine"

throw310822 — Wed, 12 Aug 2026 22:50:39 +0000

> The facility runs on 100% renewable energy and uses zero water for cooling — relying entirely on free cooling from the mine's constant underground ambient temperatures.

I'm sure they made their calculations but I wonder where do they expect 6MW of generated heat to go. Does the underground environment remain at the same temperature despite heating? Where is the heat going?

New comment by throw310822 in "Learning more about Claude's mathematical capabilities"

throw310822 — Mon, 10 Aug 2026 20:25:45 +0000

Indeed, that's a paragraph straight out of Lem's Cyberiad.

New comment by throw310822 in "Tom Stanton's supersonic trebuchet breaks sound barrier with gravity alone"

throw310822 — Sun, 09 Aug 2026 20:14:11 +0000

But since potential energy from gravity is mgh and kinetic energy of a projectile is 1/2 mv^2, calculating height and mass of a weight to power the projectile is absolutely straightforward. The terminal velocity of a falling weight has nothing to do with it (even historical trebuchets used a leveraged system in any case).

New comment by throw310822 in "Tom Stanton's supersonic trebuchet breaks sound barrier with gravity alone"

throw310822 — Sun, 09 Aug 2026 20:03:48 +0000

Of course, but there is a continuum between "I spent years tackling this complicated problem and the solution will benefit us all" and "I've entered the same McDonald's drive-through 1000 times in one day to order always the same thing". Some achievements are closer to the first end, others lie more somewhere in between.

New comment by throw310822 in "Tom Stanton's supersonic trebuchet breaks sound barrier with gravity alone"

throw310822 — Sun, 09 Aug 2026 19:59:13 +0000

No, I understand, it's the claim from the very first sentence that "gravity itself seemed to forbid" (reiterated later in the text) that annoyed me. It's just making up a non-existent obstacle to hype up the result.

New comment by throw310822 in "Tom Stanton's supersonic trebuchet breaks sound barrier with gravity alone"

throw310822 — Sun, 09 Aug 2026 19:18:15 +0000

What's extraordinary about this? You can get any amount of energy from gravity and transferring it to a small object doesn't sound like an especially complex feat. It's one of those influencer videos in which someone devotes a disproportionate amount of effort to accomplish something pointless- Mr. Beast does the exact same without any pretense of showing something useful.

New comment by throw310822 in "US Military's cyber command unit grapples with cluster of deaths by suicide"

throw310822 — Sat, 08 Aug 2026 19:44:48 +0000

No they shouldn't because "death to" is simply a mistranslation. "Death to" is an idiom in Persian that means approximately "down with"- and in fact the official translation by the Iranian government itself is "Down with the USA". But I'm curious what you thought it meant.

New comment by throw310822 in "US Military's cyber command unit grapples with cluster of deaths by suicide"

throw310822 — Sat, 08 Aug 2026 19:26:41 +0000

"Bringing democracy" is a very weak ideological motivation. I would rather call it plausible deniability. As in "we're doing this to bring democracy" and people might be left with the doubt that it's true.

New comment by throw310822 in "DeepMind's WeatherNext model achieves breakthrough forecasting cyclones"

throw310822 — Sat, 08 Aug 2026 14:08:48 +0000

Next step: steering them. (As in Permutation City's "Operation Butterfly".)

New comment by throw310822 in "I am retiring from fulltime writing (& pseudonymity) to launch Guardian Angel"

throw310822 — Wed, 05 Aug 2026 07:36:17 +0000

An AI that thinks like its user seems an awful idea to me. It's exactly the sycophancy issue that everyone has been complaining about- a user that suffers of delusions or any sort of crackpot ideas can get them reinforced by such an AI. It would be the opposite of helpful. One of the most valuable aspects of an AI partner is its otherness, the ability to provide some external or objective point of view. If it's just repeating yours, it's useless.

AIs are already, to a certain extent, ego-less: they are quick to adopt the point of view of their user. And when their intelligence is- except in case of overt censorship and political biases- always used to advance the interest of their user.

Finally the idea that a user-aligned AI would make its human user irreplaceable by companies seems a total non-sequitur, or I didn't get the argument at all (in which case maybe someone can explain it).

New comment by throw310822 in "The end of the EU is near, and it's going to get ugly"

throw310822 — Tue, 04 Aug 2026 10:49:53 +0000

> keeping von der Leyen in her current role has to be one of Moscow’s top strategic priorities. Who else is more qualified to destroy the European Union (EU) from within?

As much as I don't like von der Leyen, I don't think this is true. The current EU leadership, vdL on top, has bet everything on supporting Ukraine against Russia, whatever the costs for the EU. And I don't think Russia is interested in a weak EU either- which probably means more subservience to the geopolitical goals of the US.

New comment by throw310822 in "Don't be a meat proxy"

throw310822 — Mon, 03 Aug 2026 16:48:59 +0000

But then we agree. Some humans will behave that way. Then direct your anger against those who do so, instead of against those (the majority) who ask the questions, read and evaluate the answers, and forward them to you only if they think they might be correct and helpful.

New comment by throw310822 in "Don't be a meat proxy"

throw310822 — Mon, 03 Aug 2026 15:58:09 +0000

You know what? If you post a question and I get a good answer from Claude, it means you haven't even been able to do the basic job of asking yourself. Not only you weren't able, you also expect me to reformat nicely the answer so you don't feel stupid reading the LLM text you should have generated yourself?

New comment by throw310822 in "Don't be a meat proxy"

throw310822 — Mon, 03 Aug 2026 15:49:58 +0000

> Not every question can be answered accurately by Claude

You are assuming that people that send you a Claude answer would send you any Claude answer. This is your assumption- basically a negative prejudice towards those who use LLMs- and has nothing to do with sending a Claude answer per se. Do you understand the difference?

New comment by throw310822 in "Don't be a meat proxy"

throw310822 — Mon, 03 Aug 2026 13:31:12 +0000

Why would I reformulate what Claude already explained in a way that I find satisfactory? Only to give you the impression that I came up with it myself?

> if you give an answer, own it. Don't throw slop grenades.

If I send you some text copied and pasted from Claude, it means that I have read it and I think it's worth for you reading it. In this sense I'm already owning it. The problem is not the pasted answer, it's your assumption that a Claude-generated text must be "slop" that I haven't read.

New comment by throw310822 in "Don't be a meat proxy"

throw310822 — Mon, 03 Aug 2026 10:16:04 +0000

No, it was a suitably nasty reply to someone who boasted of being nasty to colleagues trying to help.

If your question can be answered by Claude, and you didn't ask Claude yourself, and when other people do it for you, you complain that you could have done it yourself, then my honest reply is that you should just shut up and learn to use Claude.