Nov 1, 2017

Axios Future

By Bryan Walsh
Bryan Walsh

Welcome back. Please invite your friends and colleagues to join the conversation. Tell me anything on your mind, including what you think about what you are reading here and in the daily stream. Just reply to this email, or reach me at steve@axios.com.

And a special note for readers in Philly:

Future Shapers: A New Era in Cancer Care: Vice President and Dr. Biden, along with Sean Parker, are joining Axios' Mike Allen on Wednesday, November 8th, to discuss the future of cancer care and you're invited. Mike will also explore this topic with the Celgene CEO Mark Alles, cancer survivor and advocate Stefanie Joho, and Elizabeth Jaffee, the deputy director of the Kimmel Cancer Center at Johns Hopkins. The location is the National Constitution Center. RSVP here.

Now, let's start with ...

1 big thing: The biggest problem in self-driving cars

The biggest difficulty in self-driving cars is not batteries, fearful drivers, or expensive sensors, but what's known as the "trolley problem." It is a debate over who is to die and who should be saved if an autonomously driven vehicle ends up with such a horrible choice on the road. And short of that, how will robotic vehicles navigate the countless other ethical decisions, small and large, executed by drivers as a matter of course?

In a paper, researchers at Carnegie Mellon and MIT propose a model that uses artificial intelligence and crowd sourcing to automate ethical decisions in self-driving cars. "In an emergency, how do you prioritize?" study author Ariel Procaccia, a professor at Carnegie Mellon, tells Axios.

The bottom line: The CMU-MIT model is only a prototype at this stage. But it or something like it will have to be mastered if fully autonomous cars are to become a reality.

How they created the system: Procaccia's team used a model at MIT called the Moral Machine, in which 1.3 million people gave their ethical vote to around 13 difficult, either-or choices in trolley-like driving scenarios. In all, participants provided 18.2 million answers. The researchers used artificial intelligence to teach their system the preferences of each voter and then aggregated them.

This created a "distribution of societal preferences" — in effect the rules of ethical behavior in a car. The researchers could now ask the system any driving question that came to mind, since it "knew" the ethical way to decide; it was as though they were asking the original 1.3 million participants to vote again.

A robot election: "When the system encounters a dilemma, it essentially holds an election, by deducing the votes of the 1.3 million voters, and applying a voting rule," Procaccia said.

Read the rest of the post.

2. The case for not worrying about robots ...

Only months ago, we were warned of the robot apocalypse — runaway automation that will vaporize swaths of today's jobs, too quickly and profoundly for the economy to create replacements. More recently, we hear an industrywide defense of the robots — the argument that, as has always happened since the early days of the industrial revolution, jobs we never imagined will overcome automation, employing everyone who wants to work.

The trouble with both camps is one of forecasting everywhere: We hear an abundance of assertion by professionals with slender factual basis for their certitude.

The case for robots was made when we sat down this week with senior company executives from Deloitte, which sits in the "don't worry" camp. Eamonn Kelly, a director with Deloitte Consulting, gave the best case we've heard yet for that scenario.

Kelly argues that when technological disruption happens, it often follows three paths:

  • It will displace people.
  • It will augment what people can do.
  • It will create a new art of the possible, including new work.

For two centuries, catastrophe has been routinely forecast from new technology, but "it's never happened because No. 3 is massively bigger than No. 1," Kelly said.

... isn't easing American worries about them

As you see in the chart above, the majority of Americans polled in a new survey by Axios and SurveyMonkey are unswayed by assurances of benign robots. They said technology will eliminate more jobs than it creates over the next decade.

Why it matters: Jobs lost to globalization were an undercurrent of the 2016 presidential campaign. Those concerns could envelop tech companies in the future as Americans are already wary of tech companies and the information they have.

3. Amazon is going after FedEx

Amazon is on a roll: It now accounts for 43 cents of every dollar spent online in the U.S. Last week, an ebullient Wall Street rewarded CEO Jeff Bezos by bidding up Amazon's share price and making him the richest man in the world. But Bezos seems only to be looking for the next killing, Axios' Christopher Matthews reports.

  • What's next: Amazon's revenue formula appears to be fundamentally changing — from a reliance on e-retail and cloud services, to fulfillment and shipping for outside sellers.
  • This is highly lucrative stuff: Analysts and industry executives say new initiatives like Amazon Key and Amazon Seller Flex show the company's potential to siphon off a large chunk of the trillions of dollars spent globally on logistics and shipping.

This is a familiar pattern: As it's done with every business it's attacked, Amazon looks like it's creating state-of-the-art services for its own use before offering them to outsiders. And that is what should worry FedEx and UPS. "The logistics industry is only just now starting to wake to this threat," Zvi Schreiber, CEO of the online shipping marketplace Freightos, tells Axios.

Go deeper: Read the rest of Chris' post here.

4. A shortcut to cybersecurity

With Russia, China, Iran and North Korea on the loose, experienced and knowing cybersecurity hands are among the world's most-sought-after workers. The trouble is that there are not nearly enough of them — estimates are that the U.S. alone could use 200,000 more cyber experts to protect the country's private and public computers. And half or fewer of those applying are not qualified, according to a survey by ISACA, an industry association.

What's happening: Endgame, a Virginia-based cybersecurity firm that has worked most closely with the U.S. intelligence agencies, launched Artemis, an intelligent chatbot.

Why it matters: Hyrum Anderson, Endgame's lead data scientist, says Artemis is a shortcut to closing the gap between inexperienced "Tier 1" computer analysts and top-flight but comparatively few "Tier 3" professionals, who know the field.

  • The volume of potentially malicious alerts is "staggering, so a real threat can be lost in the noise," Anderson tells Axios.
  • But by typing questions using natural English into Artemis, a relatively new cybersecurity analyst can conduct a sophisticated investigation of a vast computer system. "Our customers are trying to protect their systems with limited resources," he said.

Be smart: The yawning shortage of professionals, propelled by a wildly active hacking community — such as BadRabbit, the most recent ransomware attack — is global. There will be 3.5 million unfilled cybersecurity jobs by 2021, forecasts CyberSecurity Ventures, an industry newsletter. The forecast includes the West and other countries including India, Japan and China.

5. Worthy of your time

Inside the decline of Sears (WSJ's Suzanne Kapner)

Computers are recognizing letters like humans (Axios)

The soon-to-be retrograde job of logger (Quartz video)

Company stock soars after adding "blockchain" to name (Axios)

E-commerce takes off (The Economist)

For big tech, a threat bubbles up from the bottom (Axios)

6. 1 fun thing: Italian vs. robot motorcycle

When it comes to motorcycle racing, unlike in Chess and Go, humans are still king.

In September, Yamaha pitted a robotic motorcycle named Motobot against Italy's Valentino Rossi, the nine-time Grand Prix world champion. The contest was a single lap on a two-mile track at Thunderhill Raceway, 145 miles north of San Francisco in Willows, CA.

The outcome: Rossi 85.7 seconds. Motobot — exceeding 125 miles an hour — 117.5 seconds. Read The Drive report and watch the video.

Bryan Walsh