yen223

"It is fundamental to language modeling that every sequence of tokens is possible." This isn't true, is it? LLMs have finite number of parameters, and finite context length, surely pige…

No one argued Flash was too slow, they argued (correctly) that Flash was closed source, proprietary, and had a lot of security issues

Is there a reason why you chose to post this comment for free, without rewards, knowing full well it's going to end up in the training data of some LLM in the future?

Matt Levine pointed out in a past article that the real danger of insider trading are company insiders being incentivised to damage the company to make a quick buck.

Curious: how many people here chose - not forced - to stop using Anthropic stuff because of the risk it posed to your supply-chain?

I recall when I was a kid decades ago, being able to plug a speaker directly into the microphone jack and use it as a microphone, without any modifications whatsoever. We could do the reverse too, plu…

Positive sign that this article wasn't AI

Are they no longer worried about the supply-chain threat that Anthropic poses?

this is incredibly weird to read. once upon a time japan was notorious for its high food prices

https://stratachecks.com A site for looking up strata information for apartments in NSW, Australia

Imagine inventing reading before writing was a thing "Me understand meaning from symbols" "What are symbols?" "Dunno"

Whoever figured out writing, all those years ago. The compounded effect of having knowledge recorded for generations to come - thereby unlocking all the other things mentioned on this list - surely sh…

"A character is literally named 'Upgraydd' with creative spelling. In the future, names have become increasingly absurd — just random syllables, product names, and numbers." Upgray…

This sounds AI haha

Just once I wish they would talk about the product they actually shipped

I've always thought of CTEs as a code organisation tool, not an optimisation tool. The fact the some rdbms treats them as an optimisation fence was a bug, not a feature.

91, which to me is a prime number no matter what the maths say

This is a losing game for Pangram There's no special "humanity" quality in text. If a human can write it, there's no reason a sufficiently strong pattern matcher can't write i…

How many of you had to stop using Claude because of the Pentagon edict?

There's a lot of overlap between the "disregard this" vulnerability among LLMs and social engineering vulnerabilities among humans. The mitigations are also largely the same, i.e. limit…

https://bsky.app/profile/weiyen.net/post/3m7kenmok4c2n I did something similar. Try framing your maths question in terms of teeth…

That's fine. I'm not making a value judgement about which LLMs you should use, if any. I'm only pushing back against someone thinking "oh HN talks about Claude a lot, therefore Cla…

I suspect we're underestimating the number of users Deepseek has in Asia.

It's probably true that Anthropic's revenue is booming. But we need massive grains of salt: a) they are private and revenue numbers for private companies are hopelessly unreliable, and b) th…

WhatsApp is more popular than Slack, Teams and Telegram combined . WhatsApp has something like 2-3 billion users worldwide WhatsApp vs Slack + Teams is a bit of an apples to oranges comparison though…

Late last year was like 3 months ago. I'm bullish on Claude. It will see a surge in users, and will likely surpass Perplexity this year. However I don't think it will catch up to even Meta A…

I like Claude, but polling done on Americans late last year shows otherwise: https://epoch.ai/data/polling Overseas numbers are likely worse for Claude. Try and ask someone not i…

A lot of such cases. Claude itself had (has?) fewer users than Perplexity, let alone Meta AI, Gemini or ChatGPT