Also Open AI/ Sam admit that the concerns were quite silly in retrospect

Didn't it used to be branded as lightweight?

https://techcrunch.com/2010/09/02/google-chrome-birthday/

> I fondly remember the good old days of 2004 when I first started using Firefox as my main browser and thinking how fresh and lightweight it felt compared to the atrocity that was IE. Firefox, sadly, got bloated over the years. So far, Chrome hasn’t put on the same weight

So, yes, 15 years ago.

I'm disappointed this wasn't a joke about https://en.wikipedia.org/wiki/Universal_property

HN is normal HTML, so I think that's fair to call 'plaintext'

I write down the time I run out of tokens each day and pray my employer will pay for more

Source?

Look up independent tests of the new tokenizer for opus 4.7. I don’t think it was double, more like 140%ish of Opus 4.6.

I can’t imagine being mad that the data collection company that I work for now wants data on _me_

Really though it seems reasonable to me. They want data to train AI, and their employees are obviously a large source.

They could already track your every click. They have root on your work MacBook. Most employers do.

Could they at least have a page somewhere letting us know what we’re allowed to do today?

Closed systems are fundamentally more secure. Principle of least privilege demonstrates this

I’m slowly switching to codex simply because Claude code is closed source and I want to hack on my harness.