@ilumium @signalapp These are all bandaids and it becomes a cat-and-mouse game with phishers. What would really help (at least for organizations of any size with an admin staff or person) is federated account management. Accounts should be linked to, e.g., an internal LDAP directory or a family group. Everybody inside this org is marked green, everybody outside is different. People from other orgs that have been marked trusted get their own indicators. Individual outside accounts should be marked trusted when the owner is personally known (that's a manual step). Random accounts, however, should be clearly marked as untrusted. Wire does a bit of that. That is, allow list trusted accounts, don't try to block list bad ones.