There has been multiple attempts to criminalise end to end encryption (E2E). Currently in the news is the EU Chat Control 2.0 (though it seems it has now been withdrawn), the UK passed the Online Safety Bill and the US DoJ is "committed to developing a coherent national and international policy that encourages responsible encryption".
My understanding it that the existing legislation in Europe and most of the conversation in the US relates to scanning content pre-encryption. This has the stated goal of identifying CSAM.
Would these laws put any requirements on third party client developers or users?
While I shall try to come up with an example to illustrate my point, please not that this is a question of law. If your answer comes down to a question of fact I have probably made a mistake with this example.
My understanding is the idea of these laws is to require for example the WhatsApp or Signal client to run a government approved algorithm prior to encrypting messages, and take some action in the case of a "hit". Because of the nature of E2E this MUST happen in the client, it cannot happen in the server.
Third party clients are available for WhatsApp and Signal. The real world situation seems complicated and is largely irrelevant but it may be relevant that in the case of WhatsApp these are contractually restricted, and the Signal App is licensed under the AGPL-3.0. I take this to mean they are OK with thrid party clients, but it seems at least that they are not advisable.
One way Signal could attempt to comply with the law would be to add the required scanning in a way that could be trivially removed by a third party client. Anyone who wanted to avoid the scanning could just use that client. It seems no obvious way to stop this without requiring some server side encryption. Is there a legal solution to this in the laws?