[Expectant_Philosopher (OP)]

Attacks against businesses and individuals seem to be exploding.  As someone without in depth knowledge necessary to protect myself, my only defense is to hide in the crowd.  Could we have a system where the machine constantly monitors itself and account activity for the behavioral signature unique to an individual?  This behavioral signature would be like an online DNA fingerprint, and utilized as in a valid license against an application license server.  Higher level of security would gather a greater number of biometric data points. Initial enrollment would set a baseline, but license server would understand the signature needs to have dynamic variation, and thus constantly updates itself, learns the language of who the individual is.  When the signature is perceived as no longer valid, applications and hardware would shut down for an hour.  In an hour a valid user would have no trouble logging in manually, as they would be asked to conduct an activity only they could replicate.  If a valid user could not still log in, the user could show themselves in person to their device vendor for physical authentication.  In this manner a user could always have their device active, but never worry about defending themselves against hakers (purposely misspelled).

[TheMoon]

Unless the machine had a mind of its own, the machine itself could be hacked. One would also need a clear definition of what hacking is.  And how best it is not used in the public interest.  For example, I half read something today about a hacking group, unsatisfied with the way authorities were dealing with things, are going to solve a mystery murder with their hacking skills.  Such software may create a temporary loop-hole for third-parties to block out consent for the sharing of data.  It is the definition which is the problem, I think. Because if the software could not be manipulated, it would have a mind of its own.