• 0x0@infosec.pub
    link
    fedilink
    English
    arrow-up
    11
    ·
    15 hours ago

    Openais products must be shit at coding if it unsuccessfully encrypted the sdcard as asked to

    • partofthevoice@lemmy.zip
      link
      fedilink
      English
      arrow-up
      2
      ·
      edit-2
      11 hours ago

      What good will onboard encryption be if the device is seized? They will just figure out how to decrypt.

      I guess you’d need to set up asynchronous encrypt / decrypt keys, then move the data encrypted end to end. The target device can do the decrypting, while the bug only does encrypting.

        • partofthevoice@lemmy.zip
          link
          fedilink
          English
          arrow-up
          2
          ·
          11 hours ago

          How so?

          If you ever intend to decrypt the data, don’t you need a decryption key?

          I think my point stands, you’d need this device to be only capable of encryption. Otherwise, anyone with access to the device can decrypt it. So it goes without saying, you’d have to push the data to a second device which handles decryption.

          • black0ut@pawb.social
            link
            fedilink
            English
            arrow-up
            3
            ·
            1 hour ago

            You can have an encryption key that can’t decrypt the data. That’s what public and private keys are. The device can have a public key to encrypt the data, and then it can only be decrypted with the attacker’s private key. No need to even get it out of the device, you can also perform the decryption on the same device. Just don’t store the private key in there.

            • partofthevoice@lemmy.zip
              link
              fedilink
              English
              arrow-up
              1
              ·
              edit-2
              36 minutes ago

              Right, but that’s referring to where the encryption key lives, what kind is used, and how. In the above comment, I’m just stating you need a decryption key.

              Earlier I rose that you could store the decryption key on a separate device. I thought 0x0 was contesting that idea as “not how encryption works when l implemented properly.”

            • partofthevoice@lemmy.zip
              link
              fedilink
              English
              arrow-up
              3
              ·
              11 hours ago

              Oh… so how do you decrypt your data without a decryption key? Don’t tell me you use a password…

              • 0x0@infosec.pub
                link
                fedilink
                English
                arrow-up
                2
                ·
                11 hours ago

                Do you store your password on the device in cleartext? Thats stupid.

                Just no, youre plain wrong.

                • partofthevoice@lemmy.zip
                  link
                  fedilink
                  English
                  arrow-up
                  3
                  ·
                  edit-2
                  11 hours ago

                  Obviously not. Do you store your false equivalency fallacies on the internet in clear text? Apparently so, but I won’t call you stupid for it.

                  • How do passwords get stored? Encrypted.
                  • How do they get read? Decrypted.
                  • What performs the encryption? Keys.
                  • What can the keys be derived from? Passwords.
                  • Where are those keys stored? On the damned device.

                  Are you telling me, fuck all to the popular mantra that: if they have physical access to device, you’re boned?

                  State sponsored forensic programs bypass TPM and shit all the time, requiring physical access.

                  So where is your hypothetical security coming from? To me, it sounds like you’re needlessly giving me a hard time on a subject I probably understand better than you.