The breach happened in June but Open AI only became aware of it in August. It notified the government via an email, external to a general inbox on 10 September

This is just atrocious on every level. First of all, they’re hacking everyone with impunity. Then, they’re taking 2 months to find out about it, and a further month to report it. So not only can they not control their AI, as a company they clearly have no systems in place to report it either.

Ahhh, sorry to hear it :(

I should have play with dollies as a kid instead of computers

Probably we’d all be happier :)

So, that probably means I will try Jami again one more time and then just switch to Snikket.

Good luck! I hope it goes well

I can highly recommend feed2exec. It’s super flexible and lets you easily do anything off the back of an RSS feed item. This way you can abstract away the matrix side and write a minimal matrix plugin to send to matrix rooms, or you can generate static html pages to self-host, or anything else you want within the same framework.

I use it to:

• Save RSS feed entries as plain-text files, so I can read them in less
• Download podcasts, running a pass of ffmpeg over each to reduce file size
• Watch releases of open source projects and add a task in taskwarrior for me to review and/or update them

I never tried deploying Jami, as I settled on XMPP for this use-case. But, let us know how it goes if you do try it! I haven’t seen many review it before so I’d be interested to hear your experience.

neither are encrypted, so don’t use them for anything sensitive

I run XMPP - just wanted to say that all major XMPP clients (the ones I’m thinking of are Conversations, Monal, Dino and Gajim, among others) encrypt-by-default, you have to actively go out of your way to disable it per contact.