Hacker News: goldenarm

New comment by goldenarm in "DeepSeek V4 Pro 0813"

goldenarm — Wed, 12 Aug 2026 17:19:21 +0000

Geometric mean of all these benchmarks :

* GPT-5.6 Sol: 65.5

* Fable 5 (w/ fallback): 64.5

* Opus 5: 64.0

* DS-V4-Pro 0813: 62.5

* Kimi-K3: 62.3

* DS-V4-Flash 0731: 55.8

* GLM-5.2: 47.3

New comment by goldenarm in "Claude Code is leaking real email address as a User-Agent string in curl command"

goldenarm — Tue, 11 Aug 2026 14:36:38 +0000

I respect Anthropic for dogfooding and vibecoding their own products.

The unfortunate consequence is low quality engineering and a billion dollar product with 15k pending Github issues.

New comment by goldenarm in "Microsoft raises Xbox prices by up to 43%"

goldenarm — Sun, 02 Aug 2026 19:13:42 +0000

PS5 prices went up 35% in the past 2 years, and PC RAM went 5x. This is a global problem

New comment by goldenarm in "How Google helped destroy adoption of RSS feeds (2023)"

goldenarm — Sat, 01 Aug 2026 20:24:41 +0000

They killed greader and RSS to push Google+, which was a brilliant decision.

New comment by goldenarm in "13 Models and 4 Agents on SWE Tasks: Go, Java, Python, Rust, TS"

goldenarm — Fri, 31 Jul 2026 17:58:06 +0000

Why are models better than agents, isn't it supposed to be the opposite? I don't understand the difference and what you are measuring.

New comment by goldenarm in "Google fixed more Chrome bugs in June than over the past two years, thanks to AI"

goldenarm — Fri, 31 Jul 2026 13:39:19 +0000

Turtles with law degrees don't exist, but LLMs that increase tech debt in code bases do. Maybe recent ones got better, and it's useful to quantify the progress being made.

New comment by goldenarm in "Google fixed more Chrome bugs in June than over the past two years, thanks to AI"

goldenarm — Fri, 31 Jul 2026 10:36:59 +0000

It's not an invention it's a question. If the number is <20% that's great news for the Chrome team.

New comment by goldenarm in "Google fixed more Chrome bugs in June than over the past two years, thanks to AI"

goldenarm — Fri, 31 Jul 2026 09:54:03 +0000

The single 13yo issue is anecdata.

They obviously have the full git blame statistics but chose not to include them in the blog post, which is a bit concerning.

New comment by goldenarm in "Google fixed more Chrome bugs in June than over the past two years, thanks to AI"

goldenarm — Fri, 31 Jul 2026 09:28:12 +0000

Elephant in the room : how many of these bugs were written by LLMs in the first place ?

Because creating 100x more bugs and fixing 100x more isn't something to be proud of.

New comment by goldenarm in "“We have information that Moonshot distilled Fable for the development of K3”"

goldenarm — Wed, 22 Jul 2026 20:14:23 +0000

I have information that Anthropic distilled the internet for Fable

New comment by goldenarm in "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"

goldenarm — Tue, 21 Jul 2026 17:54:01 +0000

LLM reception is truly extreme, even worse than AAA game releases.

Ever frontier lab lived it at least once : missing the frontier by a few months triggers extremly negative reactions, then you take back the lead for 2 weeks, and the hype cycle repeats.

New comment by goldenarm in "The creepiest 'sales demo' of all time"

goldenarm — Mon, 20 Jul 2026 23:45:09 +0000

They complied by building a blocking system, instead of just disabling cookies with a flag, which would have taken exactly the same effort.

New comment by goldenarm in "The creepiest 'sales demo' of all time"

goldenarm — Mon, 20 Jul 2026 21:58:18 +0000

Why is this site blocked in the EU ? Was disabling cookies on your static page that hard ?

New comment by goldenarm in "Kimi K3: Open Frontier Intelligence"

goldenarm — Thu, 16 Jul 2026 21:22:18 +0000

All Gemma models were <30B so far, Google doesn't want to cannibalize its Gemini line too much

New comment by goldenarm in "Sony deletes more movies from the accounts of people who ‘bought’ them"

goldenarm — Thu, 16 Jul 2026 13:53:26 +0000

IANAL, but is it illegal to have a "Buy" button that is just a disguised "Rent" button?

If not, should we change the law?

New comment by goldenarm in "Show HN: Super Dario"

goldenarm — Mon, 13 Jul 2026 18:47:06 +0000

Funny idea but wow the gameplay is terrible and full of bugs. Bad hitboxes, shells that bounce in front of holes, bad physics.

AI still cannot oneshot 2D platformers, the most documented videogame genre with so much source code available.

New comment by goldenarm in "AI Boosts Research Careers but Flattens Scientific Discovery"

goldenarm — Sun, 12 Jul 2026 15:34:43 +0000

100% agree. You could make the same argument for Hollywood : funding & revenue was always the goal, and we've been producing slop before AI was even a thing

New comment by goldenarm in "EU Parliament greenlights Chat Control 1.0"

goldenarm — Thu, 09 Jul 2026 16:29:13 +0000

The US built mass surveillance by bypassing congress, at least in the EU we do it democratically /s

New comment by goldenarm in "AI builders outnumber AI governance hires 7:1 in Europe"

goldenarm — Thu, 09 Jul 2026 13:45:46 +0000

The regulations are not meant to be productive, they are meant to reduce negative externalities.

It's like minimizing the cop-to-criminal ratio. Sure it would save money, but would society come out improved ?

New comment by goldenarm in "Top researchers leave USA for the Netherlands (in Dutch)"

goldenarm — Tue, 07 Jul 2026 12:45:36 +0000

Maybe the current administration underestimates the impact of public research, and thinks Silicon Valley appeared out of nowhere.