I disagree in that I see bugs like this help bad actors. If users are conditioned to accept false alarms from rather common use cases, then that weakens true al

I disagree in that I see bugs like this help bad actors. If users are conditioned to accept false alarms from rather common use cases, then that weakens true alarms about actual stalkers. It's an inherent contradiction in the AirTag's anti-stalking requirements: it needs to be a useful alarm and not go off under "normal" proximity relationships, but most stalkers are people close to the victim to begin with. If I were building a system with such self-defeating requirements, I would take a step back and either roll it out very slowly, or conclude that this is a bad idea completely.