Hacker News: suddenlybananas

New comment by suddenlybananas in "An AI Vibe Coding Horror Story"

suddenlybananas — Tue, 14 Apr 2026 09:49:51 +0000

I don't think you read the article very carefully, the timeline is that he met a person, and that person told him that they made vibe-coded an app after having seen a video. He then investigated the app.

New comment by suddenlybananas in "System Card: Claude Mythos Preview [pdf]"

suddenlybananas — Wed, 08 Apr 2026 12:37:23 +0000

I am certain I have subjective experience.

New comment by suddenlybananas in "System Card: Claude Mythos Preview [pdf]"

suddenlybananas — Wed, 08 Apr 2026 12:30:51 +0000

OpenAI didn't want to make GPT2 available because it was "too dangerous" [1].

[1] https://www.theguardian.com/technology/2019/feb/14/elon-musk...

New comment by suddenlybananas in "Caveman: Why use many token when few token do trick"

suddenlybananas — Sun, 05 Apr 2026 11:33:20 +0000

>Some languages are far more expressive and specialized in logical conditions, conditionals, recursion and reasoning. Like eskimos have 100 words for snow, but for boolean algebra.

This is simply not true.

New comment by suddenlybananas in "German men 18-45 need military permit for extended stays abroad"

suddenlybananas — Sat, 04 Apr 2026 17:23:53 +0000

Stop reading Curtis Yarvin's pseudo-history. Like 8 million people died in the Thirty Years War before modern democratic states, and there's plenty of other examples.

New comment by suddenlybananas in "Emotion concepts and their function in a large language model"

suddenlybananas — Sat, 04 Apr 2026 17:16:37 +0000

I just don't know. I'm certain other humans are, everything beyond that I'm less certain. Monkeys wolves and rabbits, probably.

New comment by suddenlybananas in "Emotion concepts and their function in a large language model"

suddenlybananas — Sat, 04 Apr 2026 10:14:19 +0000

I'm not sure what evidence would convince me, but I don't think the way LLMs act is convincing enough. The kinds of errors they make and the fact they operate in very clear discrete chunks makes it seem hard to me to attribute them subjective experience.

New comment by suddenlybananas in "Emotion concepts and their function in a large language model"

suddenlybananas — Sat, 04 Apr 2026 08:31:16 +0000

I know I feel experience. I don't know for sure if you do, but it seems a very reasonable extension to other people. LLMs are a radical jump though that needs a greater degree of justification.

New comment by suddenlybananas in "A.I. Helped One Man (and His Brother) Build a $1.8B Company"

suddenlybananas — Thu, 02 Apr 2026 14:35:05 +0000

Yeah illegally running a cab is basically the same thing as being Rosa Parks.

New comment by suddenlybananas in "A.I. Helped One Man (and His Brother) Build a $1.8B Company"

suddenlybananas — Thu, 02 Apr 2026 12:57:35 +0000

Yes doing illegal things and getting away with it is a great way to make money.

New comment by suddenlybananas in "What major works of literature were written after age of 85? 75? 65?"

suddenlybananas — Wed, 01 Apr 2026 09:56:14 +0000

What are you referring to with his "slave daughters"?

New comment by suddenlybananas in "Model Collapse Is Happening, We Just Pretend It Isn't"

suddenlybananas — Thu, 26 Mar 2026 21:07:20 +0000

I used to think it was "modal" collapse. I still think that would be a fun alternative name.

New comment by suddenlybananas in "Buying a retro game console"

suddenlybananas — Thu, 26 Mar 2026 09:29:06 +0000

I'm surprised the author finds emulation not worth covering, that's obviously the best way to play old games in the vast majority of cases. I guess maybe they wanted to keep up the kayfabe that everyone is getting roms from the original media.

New comment by suddenlybananas in "ARC-AGI-3"

suddenlybananas — Thu, 26 Mar 2026 08:03:34 +0000

>But with LLMs, we are architecturally built the same way: it is a Neural Network that processes and makes decisions.

There are high-level similarities between ANNs and the human brain but they are very, very, very different in a ton of ways.

New comment by suddenlybananas in "ARC-AGI-3"

suddenlybananas — Thu, 26 Mar 2026 08:01:00 +0000

LLMs haven't passed us in language, a child can learn language with so so much less data than an LLM can

New comment by suddenlybananas in "Apple Just Lost Me"

suddenlybananas — Wed, 25 Mar 2026 15:05:57 +0000

More Brits have passports (86.5% [1]) than credit cards (64% [2]).

[1] https://www.ons.gov.uk/peoplepopulationandcommunity/populati...

[2] https://www.money.co.uk/credit-cards/credit-card-statistics

New comment by suddenlybananas in "The bridge to wealth is being pulled up with AI"

suddenlybananas — Tue, 24 Mar 2026 15:19:48 +0000

Well, all of the comments on the old submissions were complaining about the essay being AI slop. I haven't read the essay so I couldn't say, but that's clearly the reason why.

New comment by suddenlybananas in "Epoch confirms GPT5.4 Pro solved a frontier math open problem"

suddenlybananas — Tue, 24 Mar 2026 13:56:21 +0000

But there is no evolutionary pressure for the Poincaré conjecture, we were never optimized for that in particular, unlike these kinds of LLMs.

New comment by suddenlybananas in "Epoch confirms GPT5.4 Pro solved a frontier math open problem"

suddenlybananas — Tue, 24 Mar 2026 09:20:49 +0000

Do you think evolutionary pressures are the best explanation for why humans were able to posit the Poincaré conjecture and solve it? While our mental architecture evolved over a very long time, we still learn from miniscule amounts of data compared to LLMs.

New comment by suddenlybananas in "Mayor of Paris removed parking spaces, reduced the number of cars"

suddenlybananas — Sat, 21 Mar 2026 14:11:43 +0000

I do wonder how many cyclists in Paris are really replacing cars versus replacing metro usage. Obviously, it's still good for people to cycle as well since the metro can be insanely crowded at times, but living in Paris, my impression is that the people who cycle are the kinds who would have been unlikely to own a car in any case.