"I use Claude (an AI) to scan my posts for spelling and grammar errors."

[mahatmakanejeeves]

Reposted by Portland Frog Brigader
https://bsky.app/profile/sheilabane.bsky.social

Philip Bump
‪@pbump.com‬

I use Claude (an AI) to scan my posts for spelling and grammar errors. On the most recent post (about the East Wing, link below), I got a really amazing addendum.
www.pbump.net/o/yes-most-a...

October 30, 2025 at 11:03 AM

I use Claude (an AI) to scan my posts for spelling and grammar errors. On the most recent post (about the East Wing, link below), I got a really amazing addendum.
www.pbump.net/o/yes-most-a...

— Philip Bump (@pbump.com) 2025-10-30T15:03:57.793Z

[UpInArms]

1. ...

[muriel_volestrangler]

2. So does this mean 'Claude', from Anthropic, does not include recent text in what it works from

and has an algorithm to say "if there's no sign of this ever happening, but the style is factual, put up a warning"?

In one sense that's not bad - perhaps it's been trained on trustworthy past reports, so perhaps it's less likely to hallucinate, or assume that it must produce a story connecting all the text it's been prompted with. But it means the Trump regime can "outweird" it.

[newdeal2]

4. I don't use Claude but assume it's similar to ChatGPT

Some of the models are trained on data up to a certain point in time, but usually they will include a disclaimer to let you know.

[Hugin]

3. Yes. Overnight, some opaque and unpublished filters have been imposed on previously open...

Systems. Or so it would seem.

[PoindexterOglethorpe]

5. I use spellcheck for spelling errors.

I happen to be pretty good at things like grammar.

It also helps to reread your post before hitting the post button, something I don't always remember to do.