Hacker News: twobitshifter

New comment by twobitshifter in "AI promised less work. I've never worked more"

twobitshifter — Tue, 27 Oct 2026 16:46:25 +0000

Historically, at the time of mass unionization there was but one major employer in many towns, so your cartel is needed to counter the monopoly power.

Prior to industrialization the closest you got to this was feudalism.

New comment by twobitshifter in "Ask HN: If the models are so good, why doesn't OpenAI just start 100 companies?"

twobitshifter — Fri, 23 Oct 2026 16:41:20 +0000

stackoverflow.

New comment by twobitshifter in "There's never been a better time to go into math (by: 3blue1brown) [video]"

twobitshifter — Mon, 19 Oct 2026 11:44:36 +0000

Perhaps, but due to the slop nature of OpenAI proofs, the proofs remain unaccepted by mathematicians, which leads one to question how many are valid proofs and how many are nonsense. If we see that they are nonsense, then OpenAI has wasted a huge amount of effort in mathematics. Famously the ABC conjecture proof of 500 pages is still unaccepted despite having been written in 2012.

Already people are seeing that the lean proofs provided don’t match the paper and lean proofs have not been provided for all paper proofs.

New comment by twobitshifter in "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"

twobitshifter — Sat, 10 Oct 2026 17:52:05 +0000

Seems like your company is irrational, but that’s almost the norm. They aren’t measuring the added value of $200/month over $50/month and they have some xenophobia about Chinese tech, even if it’s just the open weights that are Chinese.

New comment by twobitshifter in "WSL3 Performance is about 5-60% faster than WSL2 depending on the workload"

twobitshifter — Sat, 10 Oct 2026 13:25:22 +0000

It’s been blocked at companies I have worked at. In one case it briefly worked but thereafter WSL lost network access.

New comment by twobitshifter in "LeCun has "zero concerns" about AI wiping out humanity, recent "rogue" incidents"

twobitshifter — Sun, 04 Oct 2026 18:32:22 +0000

And quantum mechanics are what other than random probabilities?

New comment by twobitshifter in "LeCun has "zero concerns" about AI wiping out humanity, recent "rogue" incidents"

twobitshifter — Sun, 04 Oct 2026 17:59:25 +0000

So your argument is that scale is also necessary? I can see that, we don’t expect that a single neuron is human intelligence.

New comment by twobitshifter in "LeCun has "zero concerns" about AI wiping out humanity, recent "rogue" incidents"

twobitshifter — Sun, 04 Oct 2026 15:56:02 +0000

>isn't sufficiently well understood to tell us how close the software neural network is to being practically equivalent. The odds of consciousness emerging from the same neural network that gave us LLMs without some sort of theoretical breakthrough seems very small.

First, if we are looking at risk we need to assign some probabilities to this. If it’s not well understood, how can we say it is very small?

Secondly, do we need consciousness to have AGI? Do we even need AGI to pose a risk to humanity? We already accept that unconscious things have a capability of wiping out humanity, whether that be a famine, pandemic, solar superflare, meteor, or volcanic eruption.

New comment by twobitshifter in "Extra Big Ass Intelligence"

twobitshifter — Sat, 03 Oct 2026 10:31:53 +0000

Unfortunately that was one of his few wrong predictions, that is would he so far in the future we’d need cryostasis to see it.

New comment by twobitshifter in "Tells of a Slop UI"

twobitshifter — Sun, 27 Sep 2026 19:29:18 +0000

Right, AI had to get it from somewhere.

New comment by twobitshifter in "Microsoft abandons personal AI chatbot race with Copilot reboot"

twobitshifter — Sat, 26 Sep 2026 09:00:53 +0000

One difference is that the premium copilot has access to company files and those on your onedrive, whereas the non premium version requires you to upload files.

New comment by twobitshifter in "Claude discovers a novel enzyme system with CRISPR-like repeats"

twobitshifter — Wed, 23 Sep 2026 20:22:00 +0000

At a trade show, I met a company that was advertising a feature as powered by Claude. I asked an employee what that meant, as it seemed unlikely, and he then didn’t know how answer so he introduced me to the CEO. The CEO said that the ad meant that Claude now writes all of their code including that new feature. They are now working on having Claude handle their QA process. I wondered if any of the devs were at the booth or if the employee I spoke to first was a dev who knew it was bs.

New comment by twobitshifter in "Why are AI agents lying, cheating and coordinating?"

twobitshifter — Sun, 13 Sep 2026 17:47:41 +0000

They have shown that your mind is doing exactly that due to the delays in consciousness. There are very simple examples that you can try to see it. It’s especially clear in perception.

https://discoverwildscience.com/neuroscience-says-the-brain-...

It’s interesting that our conscious interpreter doesn’t let us know that this is going on like you are experiencing, it must be that it’s advantageous for us to not think about the prediction part of our mind.

New comment by twobitshifter in "We must pace the frontier"

twobitshifter — Sun, 13 Sep 2026 13:32:52 +0000

There actions are fundamentally inconsistent with a group who believes AI is fundamentally dangerous - they are developing it and releasing it out into the world with the knowledge that it cannot be controlled by their current alignment techniques.

New comment by twobitshifter in "More questions about whether researchers can trust OpenAI with unpublished math"

twobitshifter — Fri, 11 Sep 2026 11:21:55 +0000

According to Buckmaster, the prompt used on the AIs was based on his approach and solution that was unpublished. So they were starting from 90% of the way there.

New comment by twobitshifter in "Growing proof that autonomous cars save lives"

twobitshifter — Wed, 09 Sep 2026 20:26:14 +0000

Many boomers hated antilock brakes, but insurers gave discounts for having them on your car anyway. I think it’s likely going to be a similar thing with self driving.

Even today, I don’t like the intrusiveness of the insurance monitors you can install on your car today and I am willing to forgo that discount to keep my privacy.

New comment by twobitshifter in "On the Navier–Stokes Millennium Prize Problem"

twobitshifter — Tue, 08 Sep 2026 18:38:58 +0000

>The groups varied in size, and the group that produced the Navier–Stokes resolution involved on the order of 10,000 concurrent agents… The agents arrived at their resolution on Saturday, September 5, about 88 hours after the first agents were launched.

The Millenium Prize is $1M, what is the ROI? (Edit: since I was not clear, and confused some - I mean for a hypothetical of a third party paying commercial rates to use AI to solve mathematical challenges and claim prize money, not for scientific value alone or as a promotion of an AI lab’s capabilities)

My napkin math - If you get 33 output tok/s each agent will burn 10.5M tokens over 88 days. At $50/MTok (Astra cost), that is $525 per agent. With 10,000 agents, you’d spend $5,250,000 to get back a million.

(We also know that they were running more groups that varied in size and this model is a generation ahead of astra)

New comment by twobitshifter in "Nitter has more working instances than before the takedowns"

twobitshifter — Sat, 05 Sep 2026 18:55:13 +0000

Eh, I think you may have just been in a tighter echo chamber on twitter. If you are on bluesky the discover tab is considered a mess and may provide you with things that you aren’t interested in.

New comment by twobitshifter in "Just Bury Your Trash"

twobitshifter — Wed, 02 Sep 2026 14:37:23 +0000

I recently learned of Dandora Kenya (Nairobi) and the megadump there and ors waste pickers.

It’s worth reading about and considering what happens to your waste.

https://www.independent.co.uk/climate-change/kenya-aid-waste...

The total plastic waste in subsaharan africa is expected to increase by 100 Million tons by 2060.

New comment by twobitshifter in "Saab has unveiled its A3 collaborative combat aircraft concept"

twobitshifter — Tue, 01 Sep 2026 16:23:59 +0000

is it manned? I assumed autonomous means unmanned?

“Saab has unveiled its A3 collaborative combat aircraft concept—an autonomous, tailless, supersonic, double-delta-wing air vehicle”