cross-posted from: https://lemmy.zip/post/22604748
The Vision Pro uses 3D avatars on calls and for streaming. These researchers used eye tracking to work out the passwords and PINs people typed with their avatars.
Archived version: https://web.archive.org/web/20240912100207/https://www.wired.com/story/apple-vision-pro-persona-eye-tracking-spy-typing/
Asking because I’ve never had the experience: how does one write anything while wearing a VR set? Please don’t tell me it’s one-finger “Fliegender Adler” on a giant floaty image of a keyboard?
This would utterly kill the comfort, convenience, and speed of touch typing, would it not? Ahh, progress… Even in Minority Report they had (friggin’ sweet-looking!) keyboards alongside their fancy futuristic FAUI*.
^((* FAUI - flailing arms UI)^)
From the article:
Your eyes are your mouse when using the Vision Pro. When typing, you look at a virtual keyboard that hovers around, and can be moved and resized. When you’re looking at the right letter, tapping two fingers together works as a click.
So they were working backwards to determine the inputs based off of the observed eye motion.
I have a much less modern VR headset and you can definitely still type on a regular keyboard while you’re wearing it. You can’t see the keyboard though, so you need to be skilled enough to touch type. I can’t find any reliable-looking statistics on it with a quick search, but it seems like that is not a very common skill
… Like what is not a very common skill? Touch typing in general? Or doing it under VR specifically?
- The latter would be quite niche I suppose.
- The former? I cry for the current and future generations. It really is not very hard to learn, realistic to master, and incredibly useful in daily (professional and personal) life.
Flying Adler approach works and is the solution for noobs. Pros either use a real world keyboard or just look at the letter they want to type and snap 🫰 like this to type that letter. You can type pretty fast with the look’n’snap technic.
Seems like something strapped to your face would be good use case for biometrics
Yea, you don’t need to type password if you use passkeys that are unlocked using retinaID or how it is called in the vision pro.
New
fearworry unlocked…Seems like this was done by working out passwords based on figuring out where people were looking and gesturing, rather than looking directly at the keyboard.
As a person using an uncommon keyboard layout, I reckon this would make it harder to hack my typing.
IF I could even get such a layout on wherever VR system I would theoretically be using… 😬
This makes perfect sense. The only way around it would be to randomize the location of the digits/letters, and I’m sure people would throw a fit if that was the case. Still it should be an option.
Or just use a f***ing password manager?! (Unlocked with retinaID same as with faceID on iPhone)
We are not in 2010 anymore…
Or just not show people what you’re typing.
That’s what they actually did if you read the article. They don’t pass through the eyes the same when you’re on a keyboard now.
just don’t input passwords while on call/streaming? anyway, looks like it’s easy to fix, just disable avatar eye movement mapping when inputing a password, also I doubt if it’s real vector of attack, more of a proof of concept maybe?
that “virtual keyboard” sounds awful, glad the flaw was caught quickly lmao I would just use a regular keyboard while in the headset, but I suppose that doesn’t work for most people who need to look at it to type.
It does render the keyboard, arms and hands into the virtual environment. And the look’n’snap technique to write is not as bad as you may think.
Couldn’t you theoretically do the same thing by watching someone’s eye movements while they’re typing on video chat (if they’re a hunt-and-peck typist)?
you’d have to move your eyes from letter to letter, like Vision Pro users
Maybe but I’m guessing most cameras don’t have as high of a res of your pupil?