quicklywilliam
9 hours ago
Details the design and architecture of Apple Watch's new "Audio Intelligence" features. In particular, regarding Siri Recap which has been discussed elsewhere:
*How Siri Recap respects those around you* By design, Siri Recap does not create a recording, does not produce a verbatim transcript, and does not identify and attribute speakers. The output is a brief, high-level summary, comparable to notes a person might write after a conversation. There is no audible signal because no raw audio is retained, and there is no way to reconstruct the original audio from a Siri Recap or share raw audio with anyone.
quicklywilliam
9 hours ago
Want to be clear that this is not me endorsing the feature. It's clear Apple did think about the privacy/social implications of the design, but it's also clear to me that there are real and perceived problems with having automatic conversation notes. Real differences from written notes: - It was recorded and transcribed by a machine, not a human - There are time stamps and locations - You can't tell someone is using this feature (it is obvious when someone is writing notes during a conversation)
As far as perceptions goes, it feels very tone deaf of Apple to launch this feature in this way at this time. Read the fucking room guys.