Basically, the idea is that a server can refuse to serve you (or degrade your experience with captchas/heavier restrictions) unless you (your device) complete a “challenge”. This could be something like the browser (through a system API) checking some device details like
root/admin
unlocked bootloader
extensions (either bad extensions or something like an Adblock)
VPN (potentially “if you have nothing to hide you have nothing to fear”)
installed apps (Adblock via DNS like blokada,
device emulation
TPM (generate secure key to make sure device is “real”)
OS state (heavily modified?, untrusted OS?)
etc. Basically making sure the “environment” is clean and not tampered with (trusted).
The problem is with what defines a “trusted” environment. It could start at just making sure the device isn’t rooted (like Android’s Safetynet/Play Integrity check; most people don’t root their device & don’t/won’t care, also easily justifiable since it can be a security vulnerability because the device is “wide open”).
Then, like the article mentions, the device makers (Google (phones, chromebooks), Microsoft (Windows, Xbox), Apple (macOS, iOS, visionOS, etc), Meta/Facebook (Oculus), etc) could change their terms for attestation and deny approval on stricter, potentially anti-consumer criteria such as device age (forcing you to buy more things).
It’s also important to note that Google is doing this already as well. It’s almost impossible to use Google with my VPN provider as I’m slammed with 5 captchas every Google.
I don’t think sites can request attestation yet, for vpn ips it’s usually that the ip/ip block has shown “suspicious” behavior & got reported either manually or picked up by bot sensors.
(Now of course it’s also bad to let Google and friends be the arbitrator of good and bad IPs, famous for the destruction of truly self-hosted email (among other things))
Basically, the idea is that a server can refuse to serve you (or degrade your experience with captchas/heavier restrictions) unless you (your device) complete a “challenge”. This could be something like the browser (through a system API) checking some device details like
etc. Basically making sure the “environment” is clean and not tampered with (trusted).
The problem is with what defines a “trusted” environment. It could start at just making sure the device isn’t rooted (like Android’s Safetynet/Play Integrity check; most people don’t root their device & don’t/won’t care, also easily justifiable since it can be a security vulnerability because the device is “wide open”).
Then, like the article mentions, the device makers (Google (phones, chromebooks), Microsoft (Windows, Xbox), Apple (macOS, iOS, visionOS, etc), Meta/Facebook (Oculus), etc) could change their terms for attestation and deny approval on stricter, potentially anti-consumer criteria such as device age (forcing you to buy more things).
It’s also important to note that Google is doing this already as well. It’s almost impossible to use Google with my VPN provider as I’m slammed with 5 captchas every Google.
I don’t think sites can request attestation yet, for vpn ips it’s usually that the ip/ip block has shown “suspicious” behavior & got reported either manually or picked up by bot sensors.
(Now of course it’s also bad to let Google and friends be the arbitrator of good and bad IPs, famous for the destruction of truly self-hosted email (among other things))