A Public Sector a la Google/Facebook?

Posted on July 10, 2016

In the book David and Goliath, Malcolm Gladwell argues advantage new organizations (startups) have over larger, established organizations (incumbents) is the ability to adapt fast to a changing environment: New organizations are malleable, free from existing infrastructure, which keeps them more agile to test out cutting-edge solutions. Ongoing competition between old and new businesses fuel innovation and ultimately offers consumers better solutions.

No organization is larger and with more legacy systems than the public sector (see article about FED). In fact, the public sector serves all citizens and encompasses diverse services, from health care, taxes and education, etc. Unlike the private sector, the public sector faces limited competition (citizens can in rare cases change citizenship), and the level of legacy systems is orders of magnitude larger. This specific challenge inspired Venture Capitalist Tim Draper to propose a bill to split of the Californian State into smaller competing states back in 2014. His thesis was that the public sector in California could operate at unfavorable efficiency, yet people would still choose to live there because of the location. One could argue that Draper has a point, as California by itself is the 6th the largest economy in the world, is the second most expensive taxpayer state, and yet is still on the brink of bankruptcy.

Infusing the public sector with chat bots

As a thought experiment, let’s imagine a public sector, starting on a clean slate — free of legacy systems and capable of applying modern technologies. Inspired by Google’s and Facebook’s recent yearly conferences, we will explore how citizens could communicate with the public sector.

Allo, a new communication app by Google, is a mixture of AI and chatbots applied to regular messaging. In short, what makes Allo different is the users’ ability to ask questions to Google within the communication feed. Powered by technology, we’re today witnessing a convergence, in which borders between complimentary services are disappearing and unified access points gain momentum. Uber is now much more than taxis — you can also order and get food delivered — either by car or bike (UberEverything). What if communication with the public sector also could be through a unified access point? Such a solution would take care of most requests and limit communicational overhead in the public sector, thus freeing up resources to be spent on other public goods.

Would the answers be accurate?

While Google’s search engine has indexed much of the world wide web, certain questions could be answered with higher accuracy by the public sector. This is because the public sector operate services in-house such as taxation and healthcare and have access to internal records. Instead of an app for health care, an app for tax, etc., citizens would have one access point (website, mobile app, etc.) to communicate to all the different institutions. This could be achieved by a single centralized system in which the databases of different public institutions are interconnected and applied artificial intelligence. With today’s evolution within machine learning, it would be possible for such a system to understand complex questions and ask for further information when relevant. A good example is Viv, created by the former Siri team.

Asimilar concept was already possible back in 2011 when IBM’s Watson managed to beat the two most successful Jeopardy contestants on TV. This proved that algorithms were able to understand complex questions and find the right answers at a higher rate than the most fast-thinking human beings.

A two-way communication gateway

Communication could be enabled both ways: For public elections, the message application could for example send you a reminder and allow you to vote through your phone. Obviously, privacy and authentication is very important when dealing with such matters. However, the Danish Government has already implemented a simple and safe solution for authentication in their e-Boks app. When accessing the e-Boks mobile app, you can verify your identity through a finger print scan on your smartphone. By connecting all social security numbers to a finger print, you would be able to ensure that no voted is counted twice. For those who don’t know e-Boks, it’s a system used in Denmark for one-way messages (banking, tax, etc), that before were sent as letters.

Animation of a voting chat bot To show you what a use-case could look like, please see the demo created via Facebook’s Chatbot Platform. It would be interesting to observe how such a solution could impact voting percentages among millennials, which today in US is just 25% compared to a 60% nationwide average.

Direct Democracy

Utilizing mobile technology to make voting more frictionless could also be used to entirely re-shape the implementation of democracy, existing in Western Countries today. Representative democracy, also known as in-direct democracy, is known as a political system in which citizens select politicians to make decisions on their behalf. This type of democracy is often presented as the only form of democracy possible in mass societies. Some arguments made to support this includes:

  1. Citizens lack specific domain knowledge to make educated choices
  2. Public votes are costly monetary, time-wise and in infrastructure
  3. Indirect democracy is required for fast decision-making

Argument for and against direct democracy

At least 2 of the 3 arguments mentioned above wouldn’t be valid in a world in which citizens could vote from their phone. The reason is that the cost of a vote would diminish and decisions for such votes could be executed easily. A practical way to empower citizens could be to create a system similar Reddit that allows citizens to create proposals, which then can be upvoted. The most upvoted proposals would then be selected for public voting.

The strongest argument against a direct democracy is whether the average citizen is capable of identifying the right answers. Humans are known for being short-sided and emotional. The question we need to ask is whether we should put trust in a chosen few or in the wisdom of the crowd? Many claim that our political system is broken. People have lost faith. Would a system, which grants more responsibly to citizens, encourage citizens to become more knowledgeable? Experimenting is the only way to really find out.

Asa final remark, this blog post probably leaves more questions unanswered than answered. One thing is certain, transforming today’s existing web architecture in the public sector to a system described above would be an enormous project. However, starting from scratch and with a combined market cap of roughly 850 B. dollars, Facebook and Google could probably do so successfully. I’d love to hear how you think technology could change the public sector. What are your ideas?