Hacker News: stevemk14ebr

New comment by stevemk14ebr in "Education shouldn't be held for ransom"

stevemk14ebr — Wed, 04 Nov 2026 19:59:18 +0000

This is basically pedantry, they want generative AI and AI search off. If anyone is confused then you should be asking more questions to understand the desires.

New comment by stevemk14ebr in "Progress on provable inference for large language models"

stevemk14ebr — Wed, 04 Nov 2026 17:48:07 +0000

This obviously an attempt at being able to say a response came from an anthropic model vs say some other model via a model router. As for me, I will keep using my model router, I don't care to be charged more for a name brand. I don't care about the brand at all, it's a commodity.

New comment by stevemk14ebr in "Genie 4"

stevemk14ebr — Tue, 27 Oct 2026 18:15:15 +0000

So what, the world model failed to model physics. The thing they said it did elsewhere. It's a valid critique

New comment by stevemk14ebr in "AI agents aren't 'going rogue.' Frontier labs are skipping security basics"

stevemk14ebr — Wed, 14 Oct 2026 19:52:26 +0000

It is true, and it can be true, if people hold other people to account. Accountability is the issue, stop normalizing it, your message normalizes it.

New comment by stevemk14ebr in "Dust: Pretraining Transformers Without Backpropagation"

stevemk14ebr — Tue, 06 Oct 2026 03:22:20 +0000

Pareto frontier the Pareto frontier and then with a little Pareto frontier you might get to the Pareto frontier

New comment by stevemk14ebr in "Pi.dev: You Said No MCP"

stevemk14ebr — Wed, 30 Sep 2026 14:02:51 +0000

If you don't know MCP by now that's between you and google

New comment by stevemk14ebr in "Every Household in This Rural Town Receives $10k If a Data Center Gets Built"

stevemk14ebr — Mon, 28 Sep 2026 02:00:36 +0000

That's for the recipient to decide

New comment by stevemk14ebr in "I vibed a proof of Conway's conjecture"

stevemk14ebr — Fri, 18 Sep 2026 17:00:30 +0000

Yea but that was a person who actually put in work and had to think about and understand the problem. They didn't just generate something with a magic box.

New comment by stevemk14ebr in "Nvidia announces native GPU programming in Rust"

stevemk14ebr — Thu, 17 Sep 2026 01:00:28 +0000

only reliable way to know how things work is to reverse engineer them

New comment by stevemk14ebr in "China's Regulators Take Aim at "AI Boyfriends""

stevemk14ebr — Mon, 14 Sep 2026 17:18:26 +0000

Yes, it already exists and it's called dating apps

New comment by stevemk14ebr in "Navier-Stokes – Tristan Buckmaster [pdf]"

stevemk14ebr — Wed, 09 Sep 2026 02:57:43 +0000

Yes. Quite simply yes. It's unethical. And it's a dick move

New comment by stevemk14ebr in "Remote workers report the highest well-being in study of 7,700 employees"

stevemk14ebr — Thu, 20 Aug 2026 03:00:41 +0000

you sound like a miserable person to be stuck in an office with

New comment by stevemk14ebr in "US strikes $1.2B deal to pay German firm to halt offshore wind projects"

stevemk14ebr — Fri, 07 Aug 2026 14:46:03 +0000

Perhaps you just dislike facts then

New comment by stevemk14ebr in "Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs"

stevemk14ebr — Wed, 05 Aug 2026 17:59:19 +0000

You're crazy of you think Demis is CEO of alphabet material. He's an accomplished AI researcher, Google does about 10000 other things than AI.

New comment by stevemk14ebr in "John Deere owners will get the right to repair equipment under FTC settlement"

stevemk14ebr — Thu, 09 Jul 2026 02:51:17 +0000

I wouldn't do it, your point is now disproven

New comment by stevemk14ebr in "Streaming services' obnoxiously loud ads become illegal on July 1 in California"

stevemk14ebr — Sat, 27 Jun 2026 23:14:24 +0000

You can do volume normalization, it's a thing. Lets not pretend this is some black magic.

New comment by stevemk14ebr in "Elevated error rate across multiple models"

stevemk14ebr — Tue, 23 Jun 2026 14:50:04 +0000

I'm sure they'll get right on that for ya bud

New comment by stevemk14ebr in "GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2"

stevemk14ebr — Sat, 20 Jun 2026 14:58:30 +0000

An LLM is not thinking, assuming and relating it to thought and universal truths is nonsense.

New comment by stevemk14ebr in "Citing Gandalf, Pope Leo says we must "disarm" AI"

stevemk14ebr — Tue, 26 May 2026 00:20:19 +0000

I wish you'd stay in your lane, his comments were awesome and basically the job

New comment by stevemk14ebr in "Trump fires NSF's oversight board"

stevemk14ebr — Sun, 26 Apr 2026 00:06:36 +0000

> It’s just a foundation? Please just let me know

We are each responsible for learning ourselves, and we live in a time where that is easier than ever. I find it odd your default position is to assume it is not important.