I’m not sure I agree that relying on keyword alert features alone is really as effective as you think. These systems always claim to pick up “casual or coded” language, but from everything I’ve read, they’re heavily reliant on preset keyword lists and basic matching. Most kids—or clever scammers, for that matter—adapt their language to avoid detection pretty quickly.
Honestly, when you say “less obvious phrases” get picked up, do you actually have examples? It seems like unless you’re continually updating the filters, really nuanced bullying, scam attempts, or anything vaguely off-script would still slip through. As for conversations with kids, “making it not a lecture” works, but isn’t there still a power imbalance you can never really bridge if devices are always being monitored? Here’s what I think is missing: There needs to be acknowledgment that none of these solutions are remotely foolproof, and sometimes this surveillance just teaches them to be sneakier, not safer. Do you genuinely think these apps provide real peace of mind?