I’ve recently discovered a set of applications that I think may be worthy of attention. They are listed as “The Pony Family,” and I have been using PGPony for a couple of weeks now. Other than as a user, I have no affiliation with these products, am not paid to endorse them and, it probably goes without saying, have a fairly limited knowledge of most things security related, so I defer to you all to be better judges.
I’m pretty certain that the author used AI signiciantly to create them (they keep churning out new apps extremely fast, and the website gave me super strong AI vibes last I checked). I’ve been playing around with PGPony, and it generally seems to work fine for the limited things I used OpenKeychain for, but I’d be very hesitant to trust it for anything serious.
Thank you for your input. A couple of follow-up questions, as my knowledge as to AI coding is practically non-existent: 1.
What, exactly, does that mean? Mail encryption? File encryption?
- Generally speaking, is there anything necessarily unsafe/insecure/not-private with using an AI coded application? or are we talking “philosophically speaking?”
You would be correct to think that as the readme states
This app is developed solo and with heavy use of AI assistance, which I don’t hide.
A solo dev with 35+ repos spanning Android/Kotlin, iOS/Swift, Desktop/Rust, PHP relays, C, Python all created/updated inside a ~2-month window would be a dead give away if the AI usage was not admitted too.
Ah, looks like they added that recently. I remember checking a while back and it didn’t say anything about AI usage.
Given how many apps they’re churning out in such a short timespan, I find it very unlikely they’re adequately reviewing everything. I would be very unsurprised if there are security issues. I frankly wouldn’t use it for untrusted input. However, if you trust whatever you’re putting into it (as in, you trust the sender of the email, source of the file, etc.), it’s probably fine.