In the previous Gmail litigation a couple of years ago Google Apps for Education users were one of the classes. In that case Koh dismissed the 1967 California

In the previous Gmail litigation a couple of years ago Google Apps for Education users were one of the classes. In that case Koh dismissed the 1967 California Invasion of Privacy Act eavesdropping claims because users' lawyers could not show the emails were confidential (see Section 632). However she didn't dismiss the wiretapping/interception claims (see Section 631). Another element required for an eavesdropping claim is lack of consent. How are you going to show that the granting of consent was in question for each and every user and was thus a "common question"? Maybe some users read the terms and other did not. Maybe some users paid attention to news reports about email scanning and others did not. See Koh's Opinion. For those interested search "paulhastings.com" "in re google". There is a copy of the unreported Opinion, with some interesting redactions on the devices Google uses to scan email. Content Onebox, Medley Server, Changeling, CAT2 Mixer, ICEbox Server, etc. What if users started putting "CONFIDENTIAL", as between sender and recipient, at the the top of all their emails? From there one could argue confidentiality implies lack of consent to eavesdropping. Apparently Google argued users understand that interceptions are part of how all email, not just Gmail, is transmitted. Intermediary SMTP servers might be "interception" but it seems like a reasonable judge could conclude if a message intercepted says "CONFIDENTIAL" then there's no permission to read it, whether the reading is done by a human or a machine programmed by a human.