[JayNeely]

So basically, I install a browser extension, do social sign in so you get my friends list, and now any time I go to a page not previously in my browser history, it's anonymously added to the feed of any of my friends who also have the browser extension installed? And the hope is this stream of passively broadcast stuff from your friends is more interesting than Facebook or Twitter's streams of stuff your friends have curated and chosen to share?

Ignoring all privacy issues and ways to deanonymize, one big area of risk I see for you is in your ability to filter content. One of the main benefits of people actively sharing is that they tend to add context as they do it. They comment on it, add tags, editorialize the title, and even based on where they're sharing it you can assume a little more about the content.

Without all of that, your only option for "applying filters" and determining "type of content user wants" is to do some kind of machine analysis. How reliably are you really going to be able to provide that? Especially for images, which on platforms like tumblr and imgur are often shared without any context (even titles or tags).

[SITZ]

I agree that getting all this browsing data from users will rather decrease the signal-to-noise ratio early on. But, that simultaneously provides a way to explore much more content on web that remains simply unknown to most people because of a lack of discovery medium. From personal experience, Most people share a mere fraction of stuff they discover online. So, even though that might find some rare gem on web, they will share only items which they would be rather associated with in contrast variety of things they might be interested in.

Although, I haven't ironed out exact details, IIRC, Deep learning techniques have improved significantly to make sense out of data if fed a huge chunk of it. Basic Video and Image comprehension techniques are pretty advanced too. I'd like to believe that this system will evolve to serve the purpose too.