Inside AI - November 17th, 2019

Inside AI (Nov 17th, 2019)

Inside AI Weekend Commentary: How Are My AI Predictions for 2019 Doing?

Subscribe | View in browser

Happy Sunday and welcome to the weekend commentary edition of InsideAI.  I'm Rob May, and every Sunday I try to write something about what I've learned investing in 70+ startups, mostly AI companies, and watching their challenges, struggles, and successes.  My goal is to highlight something that makes you think.  

Let's get started by reviewing the most popular articles of the past week:

Digital rights activists in Washington D.C. scanned the faces of thousands of people with Amazon’s Rekognition software on Thursday to demonstrate the harmful consequences of permitting facial recognition surveillance. The activists navigated the nation's capital in white jumpsuits with cell phones strapped to their heads to scan people they encountered outside the halls of Congress and inside the city’s busiest metro stations. As they collected their data, digital rights group Fight for the Future cross-checked the scanned faces with a database of journalists, lobbyists and members of Congress. The unique form of protest was intended to show these groups of people how this technology could affect them directly. - VICE

John Carmack, the CTO of Facebook’s VR subsidiary Oculus, is stepping down from the full-time position to focus on artificial general intelligence, also known as strong AI. In a Facebook post, Carmack said he will work from home on his AI efforts but stay on as a consulting CTO at Oculus VR. As far as AI, he said, "I think it is possible, enormously valuable, and that I have a non-negligible chance of making a difference there, so by a Pascal’s Mugging sort of logic, I should be working on it." Strong AI is considered the more "science fiction" version of AI, with a goal of teaching machines to exhibit human-level intelligence. - TECHCRUNCH

Canadian authorities have denied visas to 24 prominent AI experts from Africa and South America, preventing them from attending an industry conference in Vancouver. Organizers of next month's Neural Information Processing Systems conference are working to have the denials overturned, according to Katherine Heller, a conference co-chair. "It is very significant for the field of AI that all voices be heard," she said. The researchers, some of whom already booked flights, were scheduled to attend a Black in AI workshop at the conference. The situation has triggered an outcry from industry experts, who argued that such denials are becoming a systemic problem in Canada and prevent diversity in the field. - CNN

The computational power required to train AI is rising seven times faster than previous rates, according to an updated analysis by OpenAI. The research lab added new data to its previous 2018 analysis, which found that the power used to train large AI models doubled every 3.4 months since 2012. New data shows that the present doubling time is more than seven times the rate that occurred from 1959 to 2012, when the amount of needed power doubled every two years, which is in accordance with Moore’s Law. - MIT TECH REVIEW

Last December I posted my 2019 AI predictions.  It's always good to check in and see if I'm doing well or not.  Maybe it turns out I'm an idiot and you want to stop subscribing.  Or maybe I'm better at predicting these trends than most.  I'll let you decide.  My 5 predictions for 2019 are below, with my commentary on their accuracy.

  1. An AI hardware company will breakout, driving a plethora of new AI hardware startups and investment. Comments:  Mostly true.  The biggest announcement of the year was probably Cerebras, but, lots of other AI chip news and investment.
  2. Automation, predictive analytics, machine vision, and chatbot areas of AI will start to consolidate, as you have lots of companies with a bit of traction but not enough to raise more funding. They will get rolled up into other startups and some big companies. Comments:  Hasn't really happened.  Some acquihires have been done but no broader consolidation.
  3. Two new jobs will grow. First — “trainers” or “data annotator” have been a small thing for a few years. They will become a big thing. Secondly, you will start to hear about “knowledge mechanics.” These are people who don’t do a process but understand how to fix it when a machine screws it up. Think of a washing machine. We don’t wash clothes by hand anymore, and most of us don’t know how a washing machine works. But we have people who design and fix washing machines. These knowledge mechanics will design and fix applied AI processes in a similar way.  Comments:  Partially true.  We've seen a rise in trainers and data annotators, both within companies and as new services, but "knowledge mechanic" has yet to show up.
  4. GANs will start to show up in applications. We haven’t seen this yet, but the tech is 4 years old. It’s about time.  Comments:  Did they "start" to show up?  Yes.  I've seen several startups actively using GAN technology in their tech stacks, so it is becoming a legitimate thing.  Plus, GANs drove a lot of the tech behind fake video and audio tools of various kinds, and this was a breakout year for them.
  5. Google will come out with something new that combines deep learning and symbolic logic processing. I’ve heard Peter Norvig speak on something like this, so I assume Google is working on it, and 2018 was a little bit of a “deep learning is getting limited” year in some areas. Google will drive some new innovation here and once they do, everyone else will jump into it too.  Comments:  So far I appear to have whiffed on this one.

So in my view I got 2.5 out of 5, or 50%.  Maybe that means you should only read half the newsletters then.  But seriously, stay tuned and in a few weeks I'll make some 2020 predictions about AI.  

Thanks for reading and enjoy the rest of your weekend.


Copyright © 2020, All rights reserved.

Our mailing address is:
767 Bryant St. #203
San Francisco, CA 94107

Did someone forward this email to you? Head over to to get your very own free subscription!

You received this email because you subscribed to Inside AI. Click here to unsubscribe from Inside AI list or manage your subscriptions.

Subscribe to Inside AI