Existing trust models are one sided: as customers and citizens we are expected to authenticate ourselves to organisations, but these organisations make little or no effort to prove who they are to us.
Proving what rights and capabilities we have, proving who we are to a necessary level of authentication, is half of what we need for a trustworthy relationship - the other half is knowing that we're dealing with the right organisation.
Receiving communication that cannot be authenticated means that we have no way to prove something is real or fake. We cannot tell if this is our bank, real-estate agent, or tax office.
This makes us vulnerable to scams - we cannot tell if something is a scam or not since we have no way of authenticating communication from legitimate organisations. Expecting us to act on un-authenticated communication is encouraging risky behaviour, and the uncertainty about whether a communication is legitimate or not corrodes our trust in institutions.
This session makes the argument that we need all organisations to authenticate themselves on every communication to every customer to the extent possible.