Everything posted by CYBERING
There are two terms you should look up. The first is packers / cryptors. These take some code, hide it in some way, and wrap it in something that decodes and runs it when the application is ran. A very simple example - eval(bytes.fromhex('7072696e74282768656c6c6f20776f726c642729')) Now it would be pretty easy to recover the original script in this case. You can just copy out the stuff being passed to the eval, run it, and see the result. That's where the next concept comes in, obfuscation. This makes the code more difficult to understand. This can mean changes in variable names, control flow, using lesser known features (could we maybe move some functionality into decorators?). These can almost always be broken, but the simple idea is there, and they will generally slow someone down when analyzing your software.
I would recommend getting a VPN service. It basically sets up a tunnel between you and a server in a cloud host. If you do not get a VPN immediately, you can look into Firefox's encrypted DNS. Basically, when you visit a site it looks up the IP for the domain name you are visiting. This used to be a in plaintext and is how most people spied on what sites you were visiting. If it is encrypted, this snooping is no longer possible. Most sites use HTTPS, so the content on the page is secure. If you are visiting an HTTP site though, the landlord (if technically sophisticated) could see what site content there was. Just in general, I highly recommend Firefox's multi-account containers. It will not stop your landlord at all, but it isolates browser sessions so you are not tracked across accounts. For example, I have all my personal emails and bank stuff in one container. Then a container for all my media (reddit, hn, hak5, etc). A separate one for work. This way the trackers can not necessarily associate my gmail account with my reddit browsing.
I was going to laugh if it was just text to speech then speech to text 🙂 What could be a cool line of research would be doing adversarial training against a speech recognition neural net, find some inaudible inputs that it accepts as valid, then using those.