Heads up: I will be traveling this week, attending the Google Next conference, so expect some disruption to the usual newsletter schedule. On any given day, it may come quite late, more than once or not at all, depending on my free time and the show’s newsworthiness.
Speaking of Google Next, the theme of today’s newsletter almost seems like a prelude to that event, where artificial intelligence and machine learning will no doubt play starring roles. I’m highlighting three posts that everyone in the business of buying, selling, investing in or developing AI commercial software should read.
I don’t agree entirely with each point the authors make, but I think these posts offer are, on the whole, necessary, insightful and grounded in realism. While I’ve no doubt been bitten by the hyperbole bug a time or two during my time writing about AI, I like to think I have made a lot of these points in one way or another over the years. Mostly in the name of pointing out the promise as well as the realities of the field, especially the commercial applications of AI, however you define it.
Artificially inflated: It’s time to call BS on AI (InfoWorld):
The headline actually doesn’t do the article justice. It’s really about AI-washing, and the practice of slapping an “AI” label on things that are not, or of overpromising on the capabilities of what a particular product or algorithm can do. AI is behind some serious progress in a number of fields, including medicine: Can’t we just be happy with that?
Five AI startup predictions for 2017 (Bradford Cross):
Bradford Cross knows a lot about this space, and folks would be advised to heed what he’s saying here. But let me try to summarize: (1) bots are played out; (2) find an application for AI and own it; (3) if you’re doing a general-purpose AI solution, you had better nail it
; and (4) VC money for AI is going to become a lot harder to come by as a lot of investors take a bath.
AI’s PR problem (MIT Technology Review):
To quote the author, Jerry Kaplan: “[W]e should resist our predisposition to attribute human traits to our creations and accept these remarkable inventions for what they really are—potent tools that promise a more prosperous and comfortable future.” The latter point might be up for debate.