Citizen-centric Hyper-E government in 2030+
AI-driven analytics combined with human judgment.
Citizen-Centric Hyper-E government In 2030+
Our project’s overall aim is to propose a citizen-centric perspective on new governance models and for better understanding the changing relationships among citizens, businesses, and governments.
Especially, team Genius is focusing on building up the citizen-centric decision-making platform which is operated by Big data analysis, Artificial Intelligence and Machine learning process for contribute to open democracy.
Emerging Global Trends That Bring Hope for 2030+
In my opinion, now in 2018, we already made our first step to 2030. Some of the ideas just existed as conceptual thinking, however, when we watch news in nowadays, the future is on its way. Currently, new technologies are in an experimental stage but 12 years later these technologies are commercialised not only for tech companies but also to the public and government, and they will lead the world. Therefore, I want to point out current situations which are foundational to the future scenario.
Google I/O 2018: Machine Learning and AI ruled the roost
The hottest news nowadays maybe ‘Google’s annual I/O developer conference’ at Shoreline Amphitheater in Mountain View, California. Google focused on Android P, Artificial Intelligence, and Machine Learning. Mainly, Machine Learning and AI ruled the roost.
Google highlighted the importance of AI and machine learning in saving for lives by making predictive healthcare. In the article about ‘Google I/O 2018’ from The Indian Express, for example, AI can help clinicians by screening the retina so that they can predict the possibilities of getting a heart stroke or attack. Also, AI can predict the future condition of patients with preventative questions such as ‘Will a patient get sick again?’ so that doctors can prepare against the patient’s health condition deteriorates. (The Indian Express, 2018)
In the website ‘Economic Times’ summarised that Google AI could analyze 100,000 data points per patient, something humans cannot do, to determine if their health is likely to get worse shortly and if they may need readmission. Currently, the accuracy of prediction level is 10% better than the traditional ways and encourage clinicians to treat in advance before patients’ health deteriorate. (The Economic Times, 2018)
There’s one other point which claims our attention. Google Duplex, the most impressive one in the conference, is the new type of AI that will be part of the Google Assistant and that will talk to people in a very-natural-sounding human voice.
In google AI Blog, they explain one case study. For the hairdresser appointment example, Google made a large volume of sample hairdresser appointment calls to teach its machine learning systems how to tackle any possible conversational scenario when the Google Assistant makes these calls on behalf of consumers. At this moment, this lived up to its demo, so CEO Sundar Pichai said that they are still developing this technology. (Google AI Blog, 2018)
However, it also has ethical issues. Google Duplex acts like the real human so that AI can deceive people. Some people said it infringes people’s right to know and privacy issues.
Not only big technology companies but also the politician start to using Artificial Intelligence and Machine learning system, and these affect the democracy.
Artificial Intelligence for Democratic Innovation
The news from ‘Independent’ described the connection between Artificial Intelligence and democracy. For example, in the 2016 election, pro-Trump bots invaded Facebook pages and Twitter hashtags that Hillary Clinton supporters used for spreading contents. (Polonski, 2017)
Due to Artificial Intelligence and Big data, the political campaigns’ efficiency is maximized based on the insight elicited from machine-learning systems. Moreover, the next step is using these in election campaigns and political life.
Machine-learning systems are automatically identified patterns in data based on statistical techniques. So this highly sophisticated targeting influence people’s emotions. The prediction about susceptibility to different arguments provides that customised messages to individual voters. However, this, of course, raises ethical questions. For instance, in the article from ‘Independent’, Donald Trump’s 2016 election campaign used AI-powered technologies to manipulate citizens. As a result, some of the people claim these tools were decisive in the outcome of the vote. (Polonski, 2017)
Citizens are often affected by political information from mass media. Also, Vyacheslav W Polonski, the writer of the article, said that Artificial Intelligence and big data analysis system could help them discover the political issues of each candidate based on their interest and what they care about most. (Polonski, 2017)
Simply put, It is true that AI, Big data, and Machine learning technologies are valuable to politicians and their election campaigns. Therefore, in politics, they will use these technologies more and more. However, they must consider in using them ethically and judiciously to ensure democracy.
The future of Government 2030+
As a service designer, when we design, we should look over the future. Design-led thinking can frame other area’s problems in a new way, and that can solve challenges what we faced with. Designers tend to be empathetic to the user’s needs by understanding as to what they think and feel.
Inspiration unlocks the future; technology will catch up. (Shurina, 2015)
Design Research as a foresight method: Forecasting the future of government
To understand future, I tried to figure out what’s going on around the future government. Based on the draft scenario from FuturGov, Gov.uk blog, and itproportal website, I identified ‘The future of government in 2030+’ with five categories which are Technology, Citizens, Weak signals, Key features, and Platform. Those keywords are represented each topic and describe future scenario as well.
In the GOV.UK blog, it said, ‘Services will shape government, not the other way round.’ (Foreshew-Cain, 2016 )Based on what I found from research, we should consider citizen’s situation, types of technology, features of the platform and overall weak signals from the future.
Thinking in code. Iterating in public. (Foreshew-Cain, 2016)
Technologies & Platforms
Technologies which is in an initial stage now will be utilized in the various platforms which will be provided by the government. Moreover, these platforms will be digital by default. Hyperconnectedness allow better prediction about upcoming issues and transparency of data. AI analyses based on Big data that provided by citizens and other existing information. Furthermore, this automated system works efficiently and can save costs as well.
These platforms are preventative based on the data analyse offered by government and citizens can observe as real-time situations. GOV.UK blog said ‘Policymaking will be service design, and service design will be making policy.’ That means ideas and implementation will be closer. So sort of implementation will be an essential part of creating new ideas.
Government & Citizens
Citizens’ participation in decision making is easily possible than now. They are the owner of their data. So, sharing data with government means there’s advantages form government. Political issues are well informed and customised to the citizens with digital tools. Most customers will be able to find services that match with their needs on their digital devices or other devices that will appear in the future.
The GOV.UK.Blog said that policy making system designed and built as a framework which not as a fait accompli but allows flexibility and feedback. (Foreshew-Cain, 2016) Therefore, AI analytics can be combined with human judgment to consider ethical issues. Eventually, AI with personal assistance will contribute to access public services.
Weak signal 01: Automated system offers convenience in discover and defines problems, however, can citizens fully trust about it?
Weak signal 02: How can Machines learning system and Artificial Intelligence consider the ethical issue and prevent losing human touch from automated decision making process?
New solutions for today’s problems are not the answer for tomorrow’s challenges. Therefore, we created a new platform that could answer for the future’s challenges.
Experiment 1: Project Good — Decision Making Platform For Urban Planning
What if the citizens would be made accountable for community decisions? Can the citizens have a sustainable place in Camden?
This cube describes the future in 2030+ with four categories which are Socio-economic development, Role of technology, Situation of citizens and Role of government. All of the information is represented features of the future based on the research above. It was made up for the workshop in CSM for introducing what’s going on in the future more straightforward, more accessible and more precise.
We are focusing on the relationship between government and citizens in the future context. The government will take the facilitator roles in the decision-making process, also provide the AI analysis to the citizens so that they can be engaged in the community decisions. The citizens will be a sovereign; the personal data access can be controlled by the citizens.
Prototype: AI analysis system which is accessible to any citizens
The citizens can create the proposals about their needs, and the others can vote and leave their comment. All proposals will be created after user review. The overall AI analysis based on the citizen’s data which allowed to be accessed by the government. However, AI analysis accuracy will be affected by the citizen’s data access level.
In the workshop in CSM, we could receive various viewpoints from the policy makers, civil servants, and service designers. Here are essential feedbacks we received during the workshop.
- It would be beneficial to create and show other examples to present how our system works from starting point to the final stage, including varied stakeholders and the impact each project would have on them.
- Try to develop the way we displayed and requested data from the users of our service. We should consider the specific types of data we may need which would likely vary per project.
- To make our system friendlier for its stakeholders, it would also be good to make it more transparent by sharing information such as maintenance and other costs that would be created by each project.
- In my opinion, we could not solve the way of setting citizens’ data accessibilities and how can citizens be motivated to open their data accessibility with the government. Also, there were too many voting chances which made the process more complicated. Additionally, prevent losing human touch from automated data analysis process.
Experiment 02: Sherlock — Decision Making Platform For Open Democracy
Provocation
What if citizens were able to have access to and control of future decision making, supported by machine learning and Artificial Intelligence?
For responding to the weak signals and feedbacks from CSM workshop, we created a new platform focusing on data transparency and including ethical issues while producing results. Also, thinking about how to promote citizens allow their data to the government for more accurate results.
Introduction
Sherlock is the AI-driven platform for decision making. It will improve citizens’ experiences in the decision-making process and contribute to enhancing democracy and society.
About Sherlock
Hyper-E Government: AI + Hyper-connectedness allows better information about upcoming issues.
Our platform predicts the problems in the future based on real-time analysis system. The system uses data from the government such as demographic, geographical information, population, social phenomenon, environments, traffic conditions, etc. This platform ensures transparency. In the last stage, citizens can become aware how the system draw these conclusions with infographic video. Additionally, they received information about what kinds of data were used for analysis so that they can accept the solutions and make reasonable amendments.
Also, this platform can create two somewhat different versions of a solution and see which one works best because it allows flexibility and feedbacks. While voting and making conclusions, it reflects minorities suggestions for promoting fairness. After that, Artificial Intelligence can combine citizens’ needs and iterate and improve the better one, just killing off the one that didn’t cut. Therefore, it can deliver citizens’ voice to local communities and governments.
Prototype: AI Prediction and data transparency
“Open data is not just about transparency. It can create public and economic value.” (Hesa.ac.uk, 2018)
For understanding the entire process, we built the prototype which allows data transparency. A flashlight represents AI, and white lego box shows the data analysis process. In this box, there are transparent panels which individually illustrate different pieces of information related to the topic. After that it referrals to citizens as notifications with pieces of information about results on digital devices, and it says it is time to vote! During the vote, citizens can freely suggest their comments and AI system modifies as the best choice. This final result referral to government and they decide to go or not.
The first video above shows the process that results come out. Each small boxes on the plate revealed different data set after combine all plates finally we can see the results. And the second video is for citizens who want to vote. It describes how the platform works and what data it used.
This interface design is for citizens. Citizens can receive the notification and vote in Sherlock app.
Learnings and Iteration
Through two experiments, our team struggled with the way to motivate people to open their data, reflecting minorities’ opinion and how future technologies can apply to our service. As a result, we decided to show how it works in the system to the public for persuading them to join us.
Before designing ‘prototype 02’, we could not explain how technologies work. Therefore, we failed to persuade the experts in a workshop. Reflecting their feedback, after we fully understand how Artificial Intelligence work, Big data analysis process and machine learning principle, we could solve problems that made us stuck. Transparency was the key and will be same in the future.
Experiment 3: Data collection — Your body accessibility
You are your identity: Biometrics.
During group experiments, our team struggled with citizen’s data accessibility. Accordingly, I experimented more about data accessibility for my personal interest. This experiment is for future personal data collection.
In 2030+, citizens are the owner of their data and privacy is the key. Not only their general data can be observed but also, I think there’s an opportunity of biometrics data can be collected by the government. These kinds of data can be used in catching severe criminals, preventing our data from hacking or terror, etc. Also, it has some problems like security.
In the future, I think the more data will be needed to providing better service to the smart citizens. Accordingly, the more diverse and specific data will be needed. So, I guessed this situation in the possible future.What if, the government wants to collect citizens’ biometrics data for the public interest?
Before the survey, I explain current situations that we are already using biometrics in our life such as airport and smart devices. Moreover, also explained the future situations that there would be the opportunities to expand the use of biometrics. The participants filled out this form based on their experiences and thinking. After that, I interviewed why you can share them and why you do not want to share them.
Main Insights
- People allowed their data that already used for using smart devices or data which might be collected through surveillance cameras such as facial recognition, voice recognition, gesture recognition, retinal scan, ear scanning, fingerprint sensor and gait analysis.
- If there were benefits when they decided to share their biometrics, they could only allow visible pieces of information.
- However, the information about their inner body does not be allowed. Because they thought the situation that government wants to access their mind or to thinking feels scary.
In a nutshell, if it is necessary to collect public’s biometrics, the government should compensate for them. The thing is that data should be collected for the suitable reason such as public’s safety because biometrics is sensitive to them. In my opinion, there’s also a possibility that that data will improve the quality of public services.
Designing service for Future of government
Designing decision-making service for future government in 2030+, our work was bridge the gap between technology, government, and citizens. As a service designer who designs for the future, we should consider the role of technology and how can it be used most effectively? The following five trends will compose the future government’s services.
- Trend 01: Government as a platform
Citizens’ needs are getting complicated and diversify. The solution to the new problem will not answer for the future. Government should invent new approaches to serve the public.
“Not government that’s changed, but government that can change.” (Foreshew-Cain, 2016)
- Trend 02: Everyone is talking: Hyper-connectivity (Forbes.com, 2013)
Hyper-connectivity will change the relationship between government and citizens. Public services should be predictable and customized.
“Hyper-connectivity, the spread of social media and the increase in online personal information are key factors which will influence identity in the futue.”
— Professor Sir John Beddington (GOV.UK, 2013)
- Trend 03: Big data and Artificial Intelligence
Understanding how the technology work in the service is the critical sector of designing services. AI can improve the service delivery in government’s services. For this, the government should invest in research and promotion.
“Making Artificial Intelligence interpretable and explainable.” (Holzinger, 2017)
- Trend 04: Data transparency in Machine Learning
Due to GDPR, these topics already placed at a crucial point. The article on the website ‘We live security’ said that without transparency, machine learning process is just ‘Black box.’ (Writer and Writer, 2017) Therefore, ensure data transparency in machine learning process will contribute to prediction in the decision-making process.
“Machine learning has become a method of choice for processing large datasets and sorting samples into groups.” (Writer and Writer, 2017)
- Trend 05: Citizen centrism
Ultimately, the citizen is the core value of the government’s service. Citizen-centric government will be collaborative, Technology-driven, Agile and responsive, Driven by insight and Third-party friendly. (Business Value Exchange (BVEx), 2016)
In the end, this project aims to design service for Citizen centric Hyper-E government in 2030+. Based on the research and experiments so far, I have suggested five trends to the future government. Also, services in the future will be fully automated and technology-based, in this context, I argue that we always take into account human-centered design in the digital services.
Bibliography
Business Value Exchange (BVEx). (2016). Citizen-Centric Government: What It Looks like and How to Get There. [online] Available at: https://businessvalueexchange.com/blog/2016/12/26/citizen-centric-government-looks-like-get/ [Accessed 16 May 2018].
Forbes.com. (2018). Forbes Welcome. [online] Available at: https://www.forbes.com/sites/vivekranadive/2013/02/19/hyperconnectivity-the-future-is-now/#b5eb87630ada [Accessed 16 May 2018].
Foreshew-Cain, S. (2016). What government might look like in 2030 — Government Digital Service. [online] Gds.blog.gov.uk. Available at: https://gds.blog.gov.uk/2016/05/11/what-government-might-look-like-in-2030/ [Accessed 13 May 2018].
Google AI Blog. (2018). Google Duplex: An AI System for Accomplishing Real-World Tasks Over the Phone. [online] Available at: https://ai.googleblog.com/2018/05/duplex-ai-system-for-natural-conversation.html [Accessed 11 May 2018].
The Indian Express. (2018). Google I/O 2018: Android P, Artificial Intelligence and Machine Learning to be focus. [online] Available at: http://indianexpress.com/article/technology/tech-news-technology/google-i-o-2018-android-p-artificial-intelligence-and-machine-learning-to-be-focus-5167608/ [Accessed 11 May 2018].
The Economic Times. (2018). Google I/O Day 2 Highlights: How AI, machine learning can save more lives by making healthcare predictive. [online] Available at: https://economictimes.indiatimes.com/magazines/panache/google-i/o-day-2-highlights-healthcare-can-be-more-predictive-thanks-to-ai-machine-learning/articleshow/64103448.cms [Accessed 10 May 2018].
Holzinger, A. (2017). Transparency & Trust in Machine Learning: Making AI interpretable and explainable | hci-kdd.org. [online] hci-kdd.org. Available at: https://hci-kdd.org/2017/10/09/transparency-trust-machine-learning-making-ai-interpretable-explainable/ [Accessed 16 May 2018].
Hesa.ac.uk. (2018). How do we maintain public trust in data innovation? | HESA. [online] Available at: https://www.hesa.ac.uk/news/23-05-2016/public-trust-in-data-innovation [Accessed 15 May 2018].
GOV.UK. (2013). Notions of identity will be transformed in the next decade. [online] Available at: https://www.gov.uk/government/news/notions-of-identity-will-be-transformed-in-the-next-decade [Accessed 16 May 2018].
Polonski, V. (2018). The use of AI in politics is not going away anytime soon. [online] The Independent. Available at: https://www.independent.co.uk/news/long_reads/artificial-intelligence-democracy-elections-trump-brexit-clinton-a7883911.html [Accessed 11 May 2018].
Shurina, A. (2018). Inspiration unlocks the future. Technology will catch up.. [online] Create Yourself Today. Available at: http://www.createyourself.today/inspiration-unlocks-the-future-technology-will-catch-up/ [Accessed 12 May 2018].
Writer, G. and Writer, G. (2017). Machine-learning algorithms need transparency to comply with GDPR. [online] WeLiveSecurity. Available at: https://www.welivesecurity.com/2017/11/13/transparency-machine-learning-algorithms/ [Accessed 16 May 2018].
