It is possible that the view of what is impressive enough in computer behavior to be called intelligence changes with each decade as we adjust to what capabilities are made available in products and services.
My sense is that they would, based on a high-level take of Babbage and Lovelace's view of the potential capability of the "analytic engine". If Babbage's Tic-Tac-Toe machine had been built, I am sure that would have been regarded as machine intelligence. Nimatron (Edward Condon) may have been the first game AI, and the capability seems similar to what Babbage was envisioning. Certainly the bogus "Turk" chess-playing hoax was considered a machine intelligence.
Conventional software could connote a form of automation, and I think any form of automation, particularly where the operations are "under the covers", would have been considered a form of intelligence.
I think the current idea of only regarding "strong statistical AI" (Machine Learning) as AI is inherently flawed because of the concept of utility. Intelligence is a spectrum, being a relative measure of problem solving strength, and artificial merely connotes a thing intentionally or skillfully constructed. The Russell & Norvig definition seems to hew to this viewpoint.