The realistic scenario is company does not advertise strong encryption services, just messaging services. Then the onus is on the government to prove the messag

The realistic scenario is company does not advertise strong encryption services, just messaging services. Then the onus is on the government to prove the message went strong encrypted. But just government not being able to see their test message in the traffic they captured on servers controlled by government does not constitute a proof. A proof would require government gets their black boxes in almost every switch/router. Which I suppose is the end game.