Hacker News new | past | comments | ask | show | jobs | submit login

Where does the data to compute on come from?

Personal information will attempted to be scrubbed, but it very much is an upload of one's detailed web history to the cloud. Google alleges they wont know it's my tracks, but these federated learning systems are all powered by endless reams of data gathered from us.

That the decision of which cohort we are in can be done locally does not change the fact that that data has to come from somewhere, is gathered, en-mass, via huge bulk collections, of very specific, detailed data. It is only weasel words that it is called "not personal data"; our individuality is being harvested, tracked, modelled via these systems. That the information is free of personal identifiers does not make me feel much better about this all.

And it is something few other enterprises will ever have the capability to repeat or compete with. It mandates the web browser be powered by clouds & big data.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: