pera

Is that true? I'm actually asking because all the ones that I have seen mentioned from the arena were generated with Claude and I just checked a few random ones and they all have commits by Claude.

I don't know how much was created by LLMs but can you still call these independent implementations?

Yeah I think it is the main indicator for many people, including myself.

Obviously it would need to be sustained for a reasonable amount of time while also showing at least stable finances.

At some point AI labs will start benchmarking their latest models by the number of humans they still have working in their companies.

> I am optimistic about the future of AI at Amazon

It's scary to think that any answer but "Strongly Agree" will very likely flag you as a risk.

Is this really the society we want?

"just accept that bad things will happen to you peasant"

Wow those examples are full of verbatim copies of FOSS code (e.g. react-dom) but without any copyright/licensing notice

It's seriously mind-blowing that this is allowed and most people seem okay with it

That makes sense, but at the same time Altman has been famously called a "sociopath" by Aaron Swartz and a rapist by her own sister and yet he kept his position after the revolt a couple of years ago.

It's hard to imagine anything worse than bombing an elementary school and yet people are still using Anthropic products as if nothing has happened.

Optics don't seem to matter much?

Can some explain why would a billionaire like Altman care about the opinions expressed on HN?

Given the political power they have, particularly now with the Trump administration, whatever "optics" exist on this forum seems completely insignificant

Everything you say can and will be trained against you

My point was that we won't care about benchmarks anymore because we would see an obvious and completely unprecedent increase in productivity (and I believe it will likely come from the same people who will develope such machine).

The reason most of the conversations are focused on benchmarks is because we are still in the age of weak AI.

You don't need super-intelligence to produce at a greater rate when your tasks are parallelizable.

To each their own. Personally I will start feeling the AGI as soon as we move from chatting about benchmark results to learn that some lab just announced the discovery of tens of novel treatments for rare diseases.

Maybe I'm too boring but it seems quite pointless to have this same prediction game every time a new model is released.

I meant it in the latter way: a metaphor can be useful as a pedagogical tool to introduce new ideas, and using the source of inspiration for this idea as the metaphor itself makes perfect sense, but unfortunately our brains seem to be prone to assign other properties of the metaphor that don't actually belong to the object of study.

I imagine this happens because we tend to conflate things that are similar, or maybe because it's not entirely clear which characteristics are being mapped in the metaphor?

Yes it is a very serious problem because it confuses a lot of folks with a great deal of power like judges and policymakers.

The first book I ever read on ML (late 90s) dedicated the entire first or second chapter exploring the distinctions between artificial and biological neurons, and even talked a bit about the philosophy of modelling. I still remember thinking back then why would the authors spend so many pages on this but now I believe it was because they understood that a metaphor can be a double-edged sword.

Save that for your local mom-and-pop store: Microsoft is a multi-billion dollar corporation with enough resources to, at the very least, provide a reliable service for enterprise clients.

The "AI is using a lot of resources" excuse was maybe acceptable last year but not in Q3 2026.

Azure has a lot of GB300s:

https://blogs.nvidia.com/blog/microsoft-azure-worlds-first-g...

Yup, already mentioned in a comment below: the IPs are in that list.

They also show up in AbuseIPDB with multiple reports.

Yeah it's not just Amazon, but so far Amazon was the only one specifically looking for URLs in source code. Interestingly it ignored URLs in the fake markdown docs.

Another detail: the scraper did not attempt to access the endpoints immediately (as it did for hrefs in htmls) but it did it on the day after, twice.

Thanks, the Amazonbot IPs that accessed my honeypot are in this other list:

https://developer.amazon.com/amazonbot/searchbot-ip-addresse...

By the way, they used false user-agents.

My concern is more in relation to trying to use an endpoint from non-public source code, it seems negligent to randomly try endpoints like this

Ruled by whom? The International Court of the Northern District of California?

If you wanna look into another example search for the Abercrombie Plan in Edinburgh: it was a very ambitious urbanistic plan to "modernize" this city. For instance they proposed to demolish all of the historical Georgian and Victorian buildings in Princes Street and replace them with brutalist buildings and a motorway.

In the mid 20th century some people believed urban motorways were "progress" and wanted to build them everywhere, see for example:

https://en.wikipedia.org/wiki/Futurama_%28New_York_World%27s...

Maybe I am missing your point but "Textbooks are all you need" distilled from GPT-3.5

Sorry but your argument doesn't seem coherent: How is the cost of RL relevant here?

It would also help if you could substantiate your initial claim (i.e. "internet training data is not where frontier capabilities come from")

> Using them was allowed as fair use

That is only relevant in the US, and even there it is still not clear-cut whether the fair use doctrine applies on all these scenarios. Outside of the US the situation is also quite different: for example take a look at the recent ruling on GEMA vs OpenAI in Germany.

The reality is that the copyright issue with generative AI is very complex and reaching anything resembling a conclusion will take much more than a few opinion paragraphs from an American district judge.

-16.5% at the moment... Is the market reacting to the $20B bond offering news?

https://www.bloomberg.com/news/articles/2026-06-22/spacex-ki...

> Today we're gonna announce something a little weird and a little crazy, but also spectacular and filled with hope.

Why is that almost every LLM generated article sounds like a LinkedIn motivational post?

(this is not a rhetorical question, I would really like to know why, from all the writing styles, this is the most prevalent one)

Kissing the ring also helps - I won't be surprised if Anthropic's next model goes full MAGA

Here in Scotland it seems the desinfo campaign targeted mostly the SNP and Swinney. I guess it's hard to know how effective it was but his party lost 6 seats in last month's elections.

https://www.heraldscotland.com/news/26188090.john-swinney-ta...