Tuesday, May 26, 2020

Book Review: Burn-In, A Glimpse Into The Future Of Man-Machine Teaming

(Note:  A colleague of mine, Kelly Ivanoff, came to me a few weeks ago with a review--a really well-written review--for the new thriller by Singer and Cole called Burn-In.  I don't have a lot of guest bloggers, but I knew that SAM's audience would be interested in the book, and I told Kelly I would be happy to publish the review.  Over the next couple of weeks, Kelly got me an advance copy of the book, and I have been reading it myself (I knew 12 years of blogging would have to be good for something, someday...).  

So, who is Kelly Ivanoff and what qualifies him to comment on the future of AI, machine learning and robots?  Check this bio out:

Colonel Kelly Ivanoff presently serves at the United States Army War College.  His previous assignment was as the Executive Officer to the Director, Army Capabilities Integration Center (ARCIC), the predecessor of today’s Army Futures Command.  He’s a veteran of three combat deployments and has four years of experience specifically working future force-related efforts including concept development and force design.
Boom.  Mic drop.  Let's get to the review...Oh, and none of this is the official position of the Department of Defense or the Army.  It's all just Kelly, me, and our opinions.  Also, I'll add my two cents on the book after you're done reading what Kelly has to say.

By Kelly Ivanoff

The United States Army sees great potential in artificial intelligence and robotics to significantly impact outcomes in future combat operations.  Army General John “Mike” Murray was recently quoted in Breaking Defense, “If you’re talking about future ground combat, you’re not talking tens of thousands of sensors…We’ve got that many in Afghanistan, right now. You’re talking hundreds of thousands if not millions of sensors.” Murray later wondered, “How do you make sense of all that data for human soldiers and commanders?”  His answer:  machine learning and artificial intelligence.

Best-selling authors P.W. Singer and August Cole must have the same convictions as senior Army leaders.  Their new book, Burn-In is a riveting work of fiction, set approximately ten to fifteen years in the future, with real world, present-day implications concerning the great potential of robotics, artificial intelligence, and man-machine teaming.  They offer prophetic examples of how the military might harness and exploit the potential of these evolving technologies to improve situational understanding, “make sense of all that data,” and make better decisions.  Importantly, they vividly describe scenarios that stimulate imagination and allow consideration of challenges similar to those prioritized by General Murray and his team at Army Futures Command.

Burn-In presents the story of FBI agent Laura Keegan, a former United States Marine Corps robot handler, who is tasked to team with a robot partner to test the limits of man-machine teaming; in other words, conduct a ‘burn-in”.  Beginning with a series of controlled experiments and exercises Keegan attempts to better understand the advanced robot she’s been provided; a TAMS (tactical autonomous mobility system).  The tests are designed to explore the robot’s physical agility and its ability to learn and, as a result, improve its own capability.  The tests also challenge Agent Keegan to expand her imagination for the employment of robots and build her trust in artificial intelligence and machine autonomous operations.  The tests are halted due to a series of what seem to be unrelated disasters that inflict great damage and kill thousands of people in the national capital region.  It quickly becomes apparent the disasters were no accident.  In response, Keegan and TAMS embark on a thrilling, action-packed race to identify, locate, and stop a revenge-motivated murderer who caused the destruction.  Through this mentally and environmentally stressful period Agent Keegan overcomes her biases and comes to embrace man-machine teaming and the use of artificial intelligence in problem solving and decision making.  Ultimately, through their portrayal of this fictional story, Singer and Cole reveal numerous real-world opportunities and challenges surely inherent in our near future.  

Burn-In is much more than just a riveting story.  Singer and Cole creatively advance important concepts about the use of robotics and artificial intelligence in defense and security-related professions.  Much can be learned from their work.  Burn-In brilliantly describes example scenarios pertaining to three of the four “initial thrusts” of the Army’s newly established Artificial Intelligence Task Force; those three being Intelligence Support, Automated Threat Recognition, and Predictive Maintenance (the fourth being Human Resources / Talent Management).  The authors also provide examples related to all of the additional Areas of Interest identified in a recent call for whitepapers issued by the Army Artificial Intelligence Task Force.  Burn-In is important for the vividly described problem-centered scenarios and the conceptual solutions offered.  

Burn-In is an exceptional read and it should be a centerpiece in the library of aspiring senior military leaders, defense officials, and those involved in military modernization efforts.  Its value lies in its description of the world as it will be.  Just as the scientist and author Isaac Asimov once argued, “It is change, continuing change, inevitable change, that is the dominant factor in society today.  No sensible decision can be made any longer without taking into account not only the world as it is, but the world as it will be”.  For this reason military leaders and those engaged in the development of military technologies and operational doctrines should read this book.  It will stimulate ideas about the future operational environment and offer conceptual solutions to the inherent challenges.  Beyond the aforementioned professional reasons, read Burn-In for the sheer enjoyment of a well told story.  It will not disappoint.    

My two cents:  I like the book, too!  It reminds me of some the early work by Tom Clancy or Ralph Peters (my favorites!), and I suspect it will have that same kind of effect on military and government professionals that read it.  

Thursday, March 5, 2020

The Coronavirus Chart That Scares Me The Most

There are lots of sites that track the coronavirus, COVID-19.  One of my favorites is the one put together by Johns Hopkins.  There is lots of data there, but the chart that scares me the most is buried in the bottom right corner of the site.  The default view shows the actual number of cases reported from mainland China, from the rest of the world, and then, more hopefully, the number of people who have fully recovered.  

It's a good chart but not the one that frightens me.  You have to click the little tab that says "logarithmic" to get to the one that makes my hair a little more grey.  If you then turn off the "Mainland China" button and the "Total Recovered" button, you get the chart that sends me running for Purel and a face mask.  You can see what it looks like at the top of the page.

It shows the number of cases worldwide outside of China.  What makes it so frightening is that it is a logarithmic scale.  That means that the Y-axis doesn't increase by equal steps.  Instead, each increase represents a ten-fold increase in whatever you are measuring.  In other words, you aren't counting 1, 2, 3.  You are counting 10, 100, 1000.

If you mouse over the yellow dots you can see the dates certain milestones were hit.  For example, the world hit 100 (10 X 10) cases (plus a few) outside of China on January 29, 2020.  See the picture below:

 About 19 days later, we hit 1000 (10 X 10 X 10) cases (See below):

Then, only 13 days after that, we hit 10,000 cases (10 X 10 X 10 X 10):

Unchecked, this implies that there will likely be 100,000 cases outside of China by about March 17, 2020 and--here's the shocker--a million cases by the end of the month.  You can do the math after that.

Unchecked.  That's the operative word in the last sentence.  China got to about 80,000 cases before they managed to turn the corner.  To get there meant taking extreme measures (like closing down a city larger than New York).

It's hard for me to imagine it getting that bad, that quickly, but that's what scares me--the math don't lie.

Thursday, January 2, 2020

How To Think About The Future: A Graphic Prologue

(Note:  I have been writing bits and pieces of "How To Think About the Future" for some time now and publishing those bits and pieces here for early comments and feedback.  As I have been talking to people about it, it has become clear that there is a fundamental question that needs to be answered first:  Why learn to think about the future?

Most people don't really understand that thinking about the future is a skill that can be learned--and can be improved upon with practice.  More importantly, if you are making strategic decisions, decisions about things that are well outside your experience, or decisions under extreme uncertainty, being skilled at thinking about the future can significantly improve the quality of those decisions.  Finally, being able to think effectively about the future allows you to better communicate your thoughts to others.  You don't come across as someone who "is just guessing."    

I wanted to make this case visually (mostly just to try something new).  Randall Munroe (XKCD) and Jessica Hagy (Indexed) both do it much better of course, but a tip of the hat to them for inspiring the style below.  It is a very long post, but it is a quick read; just keep scrolling!

As always, thanks for reading!  I am very interested in your thoughts on this...)