Issue 6 | 22 March 2016

Every other week, the Fueled team hosts our own take on a book club. We call it Article Club and we all pop into a big conference room in our SoHo office to debate and discuss ideas and issues affecting tech. The basis of this conversation is 2-3 articles selected by senior members of the team.

The Articles

Capitaine Train: 10m Funding & 6 Years Later - The Story of an Unbreakable Team on a Mission - Submitted by Theo Blochet
Deep Learning Is Going to Teach Us All the Lesson of Our Lives: Jobs Are for Machines- Submitted by Perry Curac-Dahl
The revolution is Upon Us: Plant-based Foods Now Have a Voice on Capitol Hill - Submitted by Justin Cohen 

This week we turned our attention to the European tech scene. Specifically Captaine Train, the successful railway ticketing platform. Capitaine Train was born following an auspicious ruling that deregulated the monopoly on railway tickets. It was February 2009, and this group of founders had just had their latest venture, Wizzgo, shut down. Immediately following Wizzgo’s demise (we’re talking days), The French Competition Council forced France’s national railway company SNFC, to open information to other ticket providers. The founders behind Wizzgo saw a massive opportunity to penetrate a market previously shut off from third party providers.

What’s really interesting about the rise of this company is the incremental change. Unlike the popular theory that a new company must be 10x better than the competition, this second mover advantage focused solely on user experience. The result was an app and website that functioned smoother than anything on the market. Even though no deals or discounts were offered, users flocked to the clean and intuitive userface.

The goal here was to have users frequent the site whenever booking European travel but never for a long period of time. While other apps aim to have eyeballs on the screen for the most time possible, Captaine Train’s goal is efficiency. They’ve succeeded in creating a far more valuable and certainly more efficient platform than the competition with improvements to UI and UX alone.

~

Artificial Intelligence, or AI, has transitioned from something found only in sci-fi movies to a frequent news headliner. The growth surrounding AI development has been exponential, leaving many to wonder: what will happen when the robots do everything more efficiently?

Most recently, three-time Go champion Fan Hui and all-time best Go player Lee Se-dol faced off against AlphaGo, an AI, and lost (repeatedly). This is a BFD. It means researchers focusing on Deep Learning have hit milestones in development much quicker than anticipated. In chess, robots were able to quickly asses the probability of success after each move was made. There was a finite amount of moves so robotics could know with certainty what the outcome of a move would be. But Go is unique in that there are essentially an infinite amount of moves. In this case, AlphaGo had to learn from every game it played and every move made to make the best call. It required deep neural networks to provide an assumption that a given move was smart after taking into account past moves. The outcome is a more human way to play - not with certainty but with hyper educated guesses.

While the advances made by AlphaGo are newsworthy, robotic and AI-focused research is common. Robots have been taking over jobs since the 90s and we’re now moving into a period of automation that will drastically lower jobs available for humans. Typical work can be divided into routine and nonroutine, cognitive and manual. Manual routine work can be easily taught to machines (and already has in many industries) since it is unchanging. Cognitive routine is moving in a similar direction. This leaves two type of work that require more in depth thinking. Strict rules can’t be applied to these job’s ever-changing and nonroutine responsibilities. In the past it seemed like building this sort of AI was quite far in the future. But with news that AlphaGo is the reigning Go God, it is clear we’re already in the deepthinking and advanced phase of AI.

So what will the future look like with AI doing all human work? There’s the theory of basic income - each person is allocated a guaranteed income regardless of work status. This would take the place of Medicaid, food stamps, and other government programs that aid the unemployed or low-income currently. Basic income has been discussed regardless of AI development, but it certainly seems like a viable solution for a society where human jobs are scarce.

Would freedom from routine and nonroutine work lead to a semi-utopian society where we can pursue artistic endeavors and focus on passions? It does sound nice. But more likely, Fueled theorized that there would be a natural evolution within our consumer economy to realign interests and stabilize the ebb and flow of a supply and demand market.