Hacker News: lovecg

New comment by lovecg in "An OpenAI model has disproved a central conjecture in discrete geometry"

lovecg — Wed, 20 May 2026 22:14:51 +0000

I’d give humans some credit, they’re an adaptable bunch. AI won’t replace humans in the same way humans did not replace cockroaches. It’s a non-sequitur.

New comment by lovecg in "An OpenAI model has disproved a central conjecture in discrete geometry"

lovecg — Wed, 20 May 2026 22:12:05 +0000

By making it think for 100+ pages https://cdn.openai.com/pdf/1625eff6-5ac1-40d8-b1db-5d5cf925d... Regular ChatGPT users don’t have a way to do that, this is something they do internally only.

edit: apparently that’s only the _condensed summary_ of the chain of thought.

New comment by lovecg in "An OpenAI model has disproved a central conjecture in discrete geometry"

lovecg — Wed, 20 May 2026 21:59:25 +0000

The problem is the amount of data with that cutoff is really minuscule to produce anything powerful. You might be able to generate a lot of 1700s sounding data, you’d have to be careful not to introduce newer concepts or ways of thinking in that synthetic data though. A lot of modern texts talk about rates of change and the like in ways that are probably influenced by preexisting knowledge of calculus.

New comment by lovecg in "Does Employment Slow Cognitive Decline? Evidence from Labor Market Shocks"

lovecg — Mon, 04 May 2026 20:40:47 +0000

I’ve noticed some people with seemingly fulfilling hobbies stop doing them after quitting their job as well. It’s entirely possible all those hobbies are valuable precisely as something powerful to latch onto and disconnect from the day job, and seem pointless the day after quitting. Seems like you had a strong sense of identity outside of your job already before quitting. Building that could be a lot of hard work for other people (and it sometimes comes as a surprise that it even needs to be built).

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

lovecg — Thu, 09 Apr 2026 10:35:22 +0000

This could have been written almost verbatim after the printing press came out and printed pamphlets became ubiquitous.

New comment by lovecg in "Who is Satoshi Nakamoto? My quest to unmask Bitcoin's creator"

lovecg — Thu, 09 Apr 2026 01:45:28 +0000

It would be hilarious if he intentionally or accidentally lost the key, and has been trying to cash out through those Bitcoin adjacent business ventures ever since.

New comment by lovecg in "AI overly affirms users asking for personal advice"

lovecg — Sat, 28 Mar 2026 16:19:36 +0000

Let’s just hope that the people in charge of the really important decisions that affect us all approach LLM generated advice with the same wisdom.

New comment by lovecg in "Return of the Obra Dinn: spherical mapped dithering for a 1bpp first-person game"

lovecg — Fri, 20 Mar 2026 01:25:45 +0000

(2017)

New comment by lovecg in "Austin’s surge of new housing construction drove down rents"

lovecg — Thu, 19 Mar 2026 00:52:44 +0000

Don’t know why this is being downvoted, that’s exactly right. One needs only to attend a local city meeting about any smallest step towards more development to see how the voters think.

New comment by lovecg in "No, it doesn't cost Anthropic $5k per Claude Code user"

lovecg — Tue, 10 Mar 2026 05:43:13 +0000

I see quite a few:

“what X actually is”

“the X reality check”

Overuse of “real” and “genuine”:

> The real story is actually in the article. … And the real issue for Cursor … They have real "brand awareness", and they are genuinely better than the cheaper open weights models - for now at least. It's a real conundrum for them.

> … - these are genuinely massive expenses that dwarf inference costs.

This style just screams “Claude” to me.

New comment by lovecg in "The first airplane fatality"

lovecg — Tue, 10 Mar 2026 05:32:14 +0000

Definitely, you can’t get much safer mode of travel than a modern airline. There are some humorous statistics that hour for hour taking a shower is more dangerous.

Accidents generally go up as you move down the scale of regularly scheduled airlines -> charter -> private with professional crew -> private flying.

New comment by lovecg in "The first airplane fatality"

lovecg — Tue, 10 Mar 2026 05:26:07 +0000

You’d be surprised. Small aircraft crashes, if control is not lost before hitting the ground, are much more similar to car crashes than to large airliner crashes. For example, a recent “innovation” that saved a lot of lives was making shoulder belts (as opposed to lap only belts) standard.

New comment by lovecg in "No, it doesn't cost Anthropic $5k per Claude Code user"

lovecg — Tue, 10 Mar 2026 05:08:58 +0000

I can’t get past all the LLM-isms. Do people really not care about AI-slopifying their writing? It’s like learning about bad kerning, you see it everywhere.

New comment by lovecg in "Files are the interface humans and agents interact with"

lovecg — Sat, 07 Mar 2026 19:25:15 +0000

As in, you used 0 AI to write or edit this text? Or some AI? I’d like to calibrate myself.

New comment by lovecg in "Global warming has accelerated significantly"

lovecg — Fri, 06 Mar 2026 16:19:50 +0000

In an indirect way it kind of is though. At the end of the day there’s more energy trapped in the atmosphere (stronger jet streams, more frequent and stronger frontal weather)

New comment by lovecg in "Ghostty – Terminal Emulator"

lovecg — Sun, 01 Mar 2026 18:15:47 +0000

Yeah it’s been amusing to see people rediscovering tmux as well

New comment by lovecg in "I converted 2D conventional flight tracking into 3D"

lovecg — Tue, 17 Feb 2026 17:56:08 +0000

If this was a jet there’s no good technical reason for it to stay so low, but one common reason could be air traffic control congestion, you go through different sectors at 10k than you would at 30k so it’s sometimes feasible to go lower and avoid long delays.

New comment by lovecg in "GPT-5.2 derives a new result in theoretical physics"

lovecg — Fri, 13 Feb 2026 20:35:09 +0000

Let’s have some compassion, a lot of people are freaking out about their careers now and defense mechanisms are kicking in. It’s hard for a lot of people to say “actually yeah this thing can do most of my work now, and barrier of entry dropped to the ground”.

New comment by lovecg in "GPT-5.2 derives a new result in theoretical physics"

lovecg — Fri, 13 Feb 2026 20:23:39 +0000

I’m guessing, would love someone who has first hand knowledge to comment. But my guess is it’s some combination of trying many different approaches in parallel (each in a fresh context), then picking the one that works, and splitting up the task into sequential steps, where the output of one step is condensed and is used as an input to the next step (with possibly human steering between steps)

New comment by lovecg in "An AI agent published a hit piece on me"

lovecg — Thu, 12 Feb 2026 18:17:01 +0000

Surely there exists a protocol that would allow to prove that someone is human without revealing the identity?