Probably keep making more until the publishers were forced to negotiate. This method is successful. See: Uber.

If it gains sufficient traction and was found to generate equal or more revenue for publishers I could see them willing to come to the table.

The problem I see with bringing publishers to the table is: How do you distribute revenue? Pageviews? That's what you don't want papers to optimize for right now. Equally? Hardly seems fair. Who decides which publishers are on Netscoopx? What ratio of subscribers to paper subscriptions should Netscoopx maintain?

Overall the idea reminds me of Grooveshark and Aereo simultaneously.

Proxy and cache articles using a standard set of credentials from a set list of origin IPs, most likely. The traffic would appear as that of an average reader.

That sounds very likely to get you sued for copyright infringement.

Or just provide a front end for caching via archive.is and other archive sites?