The constant evolution of technology has had a profound impact on our society. The digital world is redefining how we do business, socialize, and even govern. Along with technology innovation comes policy changes. Technology and public policy are not only intertwined; they affect one another.
Technology’s influence on public policy can take many forms: from regulating what content is made available online, like net neutrality laws, to rethinking how government operates and fulfills its duties, like adaptive management and open-source governance, or by changing how we think about education.
While the evolution of technology is happening fast, policies are slow to catch up. As we learn more about how technology can be beneficial to society, we will continue to push for policies that support this vision of a mutually beneficial relationship to maximize our society's benefits. In this article, we will examine the relationship between the two a little closer.
The Relationship Between Innovation and Policy
Technology disrupts and innovates, while public policy governs and controls. As a result, the connection is complex but increasingly intertwined. At all levels of our society, we are developing sophisticated socio-technical systems of varying complexity.
Software constrains behavior in a way that no legislation can come close to. Everything is changing rapidly; technology is reshaping the reality in which we all live, and policymakers cannot keep up [1]. Maintaining separate boundaries between technology and policy restrains innovation that could benefit society as a whole.
Here are just a few examples of how technology and policy are intertwined and what this means for society:
Artificial Intelligence
AI “has the ability to supplement human decision-making, ultimately replacing notoriously subjective human processes with something more impartial, consistent, quicker, and scalable” [1]. However, it can be hacked, offering offenders ranging from criminals to national governments new tools for disruption and damage.
AI can also entrench prejudice and codify inequality and behave in inexplicable and undesired ways due to AI bias. For example, a US court's risk assessment computer algorithm was prejudiced towards black inmates. The program, Correctional Offender Management Profiling for Alternative Sanctions (COMPAS), tended to misclassify black offenders as more likely to re-offend than whites [3]. This is because human prejudices can be coded into AI and machine learning algorithms. Unfortunately, if AI bias is overlooked, it could lead to biased judgments and decisions that appear to be objective since they come from a non-human entity.
So how can we avoid the dangers of AI while reaping the benefits? This is where policies come into play. If our judgment is misplaced, it could lead to adverse situations where privacy is no longer protected. In addition, it could make biases and unfair decisions appear credible because they have been coded into technology. However, if public policies are put into place correctly, then AI can have a positive impact on personal lives and businesses.
Robotics (an AI Application) and Automation
Expert opinions differ on the extent of automation and robotics effect on the workforce. Some argue that work will be replaced by robotics and automated machines [4], and some argue that technology may generate new job categories that will hire displaced people [5]. Others believe computers will have minimal impact on future jobs. Robert D. Atkinson argues “notwithstanding some studies that suggest the next production system will lead to higher structural unemployment and reduced labor incomes, the evidence and logic suggest structural unemployment will not increase, and labor will receive a significant share of the benefits” [4]. Any policy initiatives addressing the future of employment must account for unpredictability.
If future automation technologies threaten jobs, there must be a method to provide benefits outside of work. Flexible security, sometimes referred to as Flexicurity [4], is a proposed public policy model for providing healthcare, education, and housing to the unemployed. In addition, with the possibility of chronic unemployment due to automation, expanding the Earned Income Tax Credit, establishing a guaranteed basic income, and promoting business profit-sharing are among the options that policymakers need to explore.
If policymakers can work these issues out, then robotics and automation have the ability to change the workplace entirely. Businesses could produce more for less and do it quickly. Also, there would no longer be a need for humans to be involved in dangerous manufacturing practices where workplace incidents and deaths are unfortunately common. The direction that automation takes in society truly depends on the public policies put into place.
Data Privacy
An example of an area in technology that needs oversight is data usage and privacy. Most people are already aware of social media algorithms that suggest content based on a user’s interest. While this data creates convenience in YouTube recommendations, seeing close friends' updates first on Instagram and Facebook, and finding relevant products on Amazon, data privacy is actually of significant concern because of the policies, or lack thereof, surrounding what companies can do with user data. For example, in May 2018, news reported that Amazon’s Alexa-powered Echo secretly recorded a family’s conversation and sent it to a random person on their contact list without the family’s permission [6][7].
Simply stated, there are no strong regulations protecting data privacy. The US lacks a comprehensive federal data privacy law [8]. As a consequence, most data collection products are unregulated. There are no federal privacy regulations, so many businesses may do anything they want with data. Three critical points regarding data privacy in the US are as follows: Most states allow companies to collect data without user knowledge, if data is hacked, no laws indicate when (or if) a business must notify the user, and businesses that share user data with other parties may sell or distribute it without informing the user [8].
In this scenario, public policy has been prolonged to catch up. As mentioned, no federal laws exist that protect a person’s right to data privacy, and the state laws that do exist minimally safeguard rights. Unfortunately, this has led to targeted advertising, geo-tracking, and data breaches negatively impacting people. There is no denying that data collection can offer some benefits, but most users are untrustworthy due to the lack of transparency involved. If policies are put into place, there will be a better trust system between companies that collect data and their users.
Looking Towards the Future
It’s not expected that technology innovation will slow down anytime soon, so policymakers will need to keep up. Differing opinions do exist regarding where legislation needs to be involved and how involved it needs to be. What’s essential is that policymakers, engineers, technologists, and other stakeholders work together to develop policies that prepare the global economy for the fast adoption and implementation of new ideas and are most beneficial and ethical for society as a whole.
References
The tensions between explainable AI and good public policy (brookings.edu)
Can the criminal justice system's artificial intelligence ever be truly fair? (massivesci.com)
Amazon Echo recorded conversation, sent to random person: report (cnbc.com)
The State of Consumer Data Privacy Laws in the US (And Why It Matters) | Wirecutter (nytimes.com)