Hacker News: diamondage

New comment by diamondage in "“Math 2.0” will need to value mathematical progress more holistically"

diamondage — Thu, 08 Oct 2026 08:17:26 +0000

This misses the raw advantage of a good proof. It makes conceptualization simpler. In some ways math is like a hash list of of theorems. This list makes it simpler to prove other calculations, and will always be useful, to both humans and AI models. I can see two new directions 1 - the creation of specialist theorem models; that can answer questions efficiently about one topic and 2 - we probably need to incentivize and codify ownership of theorems; charging a proportion of the compute saved by using them. Ultimately enabling mathematicians to be paid our true market value!

New comment by diamondage in "Aging brains blend memories together instead of just forgetting them"

diamondage — Thu, 03 Sep 2026 16:49:37 +0000

Perhaps undiscussed in the paper is how the blending of memories may also be better. If evolutionary pressure encourages the structure of brain to optimise for accurate prediction, not memory, then it is quite possible that older individuals may be making better predictions than younger ones, in situations where they don't remember the details...

New comment by diamondage in "Show HN: The Front Page – Newspaper-style front page for Hacker News"

diamondage — Mon, 25 May 2026 02:03:15 +0000

Its great. The article summaries frequently have odd phrases in them - eg "keep concise", the precise summary" etc. Might want to check that!

New comment by diamondage in "Decisions that eroded trust in Azure – by a former Azure Core engineer"

diamondage — Fri, 03 Apr 2026 00:59:17 +0000

In my experience Azure endpoint versus openAI endpoint was way faster and significantly cheaper.

New comment by diamondage in "The Waymo World Model"

diamondage — Fri, 06 Feb 2026 22:25:06 +0000

The uptake forced the bigger companies to act. With image diffusion models too - no corporate lawyer would let a big company release a product that allowed the customer to create any image...but when stable diffusion et al started to grow like they did...there was a specific price of not acting...and it was high enough to change boardroom decisions

New comment by diamondage in "A case study in PDF forensics: The Epstein PDFs"

diamondage — Wed, 04 Feb 2026 17:21:26 +0000

Why are they wrong? Surely it depends on how you train it?

New comment by diamondage in "A flawed paper in management science has been cited more than 6k times"

diamondage — Sun, 25 Jan 2026 12:30:33 +0000

I'm not sure if you're correct. In fact there has been a revolution in some areas of social science in the last two decades due to the availability of online behavioural data.

Code four pitchdeck published by business insider

diamondage — Fri, 21 Nov 2025 08:13:00 +0000

Article URL: https://www.businessinsider.com/code-four-mit-dropouts-arming-police-ai-pitch-deck-2025-11

Comments URL: https://news.ycombinator.com/item?id=46002337

Points: 1

# Comments: 0

New comment by diamondage in "Queueing to publish in AI and CS"

diamondage — Thu, 02 Oct 2025 19:23:00 +0000

The question 'what is science actually for' can be sidestepped. Everyone can agree that it has value, albeit we disagree on the actual value...this is why you need a market. As to how things get priced in such a market, this is a subject for further research...To start, it just needs to tie to something measurable. Heck we've created memecoins with far less backing. Also, we've carved up the conceptual space on a very course grained level with patents, we just need a more immediate, and granular system for doing so...

New comment by diamondage in "Queueing to publish in AI and CS"

diamondage — Mon, 29 Sep 2025 18:32:07 +0000

Actually, the problem is pricing. If we could identify and correctly value new concepts, then we can dispense with citations and just use the correct sum of concept valuations. Perhaps a correctly designed futures market would not only solve getting the right PhD students the right jobs, but bring a lot of speculative capital into fundamental research?

New comment by diamondage in "Novel hollow-core optical fiber transmits data faster with record low loss"

diamondage — Sat, 06 Sep 2025 07:52:39 +0000

Low latency starlink orders on hold

New comment by diamondage in "Show HN: Whispering – Open-source, local-first dictation you can trust"

diamondage — Tue, 19 Aug 2025 20:14:06 +0000

How well does it work with multiple languages?

New comment by diamondage in "The Robot Sculptors of Italy"

diamondage — Sun, 13 Jul 2025 21:31:49 +0000

Didn't they get their robots from Italy?

New comment by diamondage in "What I learned gathering nootropic ratings (2022)"

diamondage — Wed, 02 Jul 2025 18:00:21 +0000

Tried creatine?

New comment by diamondage in "Getty drops primary claim against Stable Diffusion"

diamondage — Sat, 28 Jun 2025 18:33:32 +0000

Is this significant, or does it set a precedent in anyway?

Getty drops primary claim against Stable Diffusion

diamondage — Sat, 28 Jun 2025 18:33:32 +0000

Article URL: https://www.pcgamer.com/hardware/getty-drops-primary-claims-against-stable-diffusion-in-ai-lawsuit-after-failing-to-establish-a-sufficient-connection-between-the-infringing-acts-and-the-uk-jurisdiction-for-copyright-law-to-bite/

Comments URL: https://news.ycombinator.com/item?id=44407002

Points: 4

# Comments: 1

New comment by diamondage in "For algorithms, a little memory outweighs a lot of time"

diamondage — Thu, 22 May 2025 09:41:33 +0000

Should have come to the comments first!

New comment by diamondage in "The A.I. Radiologist Will Not Be with You Soon"

diamondage — Fri, 16 May 2025 08:18:20 +0000

I did work on a diagnostic assist tool developed by a large pharma which diagnosed between Asthma and COPD. GPs get this right 52% time, specialists just over 60 and AI came in over 80%...using 12 relatively mundane input variables. I believe there are a lot of these situations, but not clear pathway to FDA approval

New comment by diamondage in "The A.I. Radiologist Will Not Be with You Soon"

diamondage — Thu, 15 May 2025 19:12:24 +0000

yes but you can train relatively dumb AI on high volumes of historical data and presenting factors. This is typically just a 'diagnostic assist', but often significantly better than a human

New comment by diamondage in "The A.I. Radiologist Will Not Be with You Soon"

diamondage — Wed, 14 May 2025 20:31:40 +0000

Underreported competitive advantage of NHS is their unified and comprehensive data