ELI Report
Author
Laura Frederick - Environmental Law Institute
Environmental Law Institute
Current Issue
Issue
3

Artificial Intelligence: Will algorithms benefit the environment? Report points the path to beneficial uses of computerization

Artificial Intelligence is changing how our society operates. AI now helps make judicial decisions, medical diagnoses, and drives cars. AI also has the potential to revolutionize how we interact with our environment. It can help improve resource use and energy efficiency and predict extreme weather.

AI can also exacerbate existing environmental issues. For example, software manipulation of over a half million VW diesel automobiles created one of the largest environmental scandals of the past decade.

ELI’s Technology, Innovation, and the Environment Program was developed to better understand the environmental impacts and opportunities created through emerging technologies and their underlying innovation systems

When Software Rules: Rule of Law in the Age of Artificial Intelligence, a new report from program director David Rejeski, explores the interaction between AI and the environment and the need for some form of governance to ensure that it is deployed in a manner that is beneficial.

“As environmental decisionmaking becomes internalized into AI algorithms, and these algorithms increasingly learn without human input, issues of transparency and accountability must be addressed,” said Rejeski. “This is a moment of opportunity for the legal, ethical, and public policy communities to ensure positive environmental outcomes.”

“When Software Rules” offers the government, businesses, and the public a number of recommendations they can use as they begin to consider the environmental impacts of AI.

The report discusses concerns with AI systems. These include unintended consequences, such as race bias in algorithms, and the common difficulty of understanding the logic of deep-learning systems and how they come to decisions. Other sources of concern include issues like algorithms functioning on the basis of correlation without proving causality; legal liability issues; lack of privacy from data mining; and the risk of hacking.

Some form of governance over AI systems is necessary to address some of these issues, and ensure responsibility, including taking environmental considerations into account. Semi-formal governance systems may include voluntary codes outlining engagement with AI research or self-governance by institutions looking to create “ethical” AI systems. A more formal governance system may include legislation protecting consumers from faulty algorithms.

ELI provides a number of recommendations as to how AI governance can include consideration of environmental impacts. Suggestions are provided for all stakeholders: the private AI sector, programmers, governments, and the public.

For example, the private AI sector can develop research teams that include evaluation of the socio-environmental impacts of their algorithms and assemble stakeholder groups to develop guidelines for sustainable development of AI.

Programmers can increase the transparency of their algorithms so users can understand why decisions are being made, and they can increase their commitment to prioritizing environmental benefits.

Governments can ensure AI systems are powered by renewable energy to meet the energy demand of these new systems and create incentives for the development of AI that tackles environmental issues.

Members of the public can advocate for systems that promote their cultural norms and values, including environmental protection, and they can make responsible consumer choices by supporting AI companies that are transparent and environmentally conscious.

As AI governance becomes a societal expectation and is later bound by semiformal or formal contracts, the environment must be a central focus in AI discourse and subsequent laws and policy, the report concludes. ELI will continue to provide guidance on how these goals can best be achieved.

“When Software Rules: Rule of Law in the Age of Artificial Intelligence” is available for free download at eli.org/research-report/when-software-rules-rule-law-age-artificial-intelligence.

Al Moumin awardees highlight promise of peacebuilding efforts

ELI co-hosted the annual Al-Moumin Distinguished Lecture on Environmental Peacebuilding, a hallmark of the Institute’s Environmental Peacebuilding Program. Co-sponsored by the Environmental Law Institute, American University, and the United Nations Environment Programme, the lecture recognizes leading thinkers who are shaping the field of environmental peacebuilding and presents the prestigious Al-Moumin Award. The series is named for Mishkat Al-Moumin, Iraq’s first Minister of Environment, a human rights and environment lawyer, and a Visiting Scholar at ELI.

This event, now in its fifth year, honored Ken Conca and Geoff Dabelko for their outstanding contributions to the field.

Conca is a professor of international relations in the School of International Service at American University. Dabelko is a professor and director of environmental studies at the Voinovich School of Leadership and Public Affairs at Ohio University; he is also a senior advisor to the Environmental Change and Security Program of the Woodrow Wilson International Center for Scholars.

Fifteen years ago, Conca and Dabelko published Environmental Peacemaking, a rejoinder to grim scenarios foreseeing environmental change as a driver of conflict. Conca, Dabelko, and collaborators argued that, despite conflict risks, shared environmental interests and cooperative action could also be a basis for building trust, establishing shared identities, and transforming conflict into cooperation.

In their lectures, Conca and Dabelko reflected on the evolution of environmental peacebuilding research since their work began in the early days of the post-Cold War era, their seminal publication, and their long-term engagement with policymakers and practitioners applying these insights around the world.

Their work transformed, and continues to have a profound impact on, the way scholars and practitioners approach and understand the intersection of environmental protection, national security, and human rights.

Conca and Dabelko’s work is also the heart of ELI’s Environmental Peacebuilding Program: As the world experiences increasing pressures on its natural resources and climate, countries must learn to peacefully resolve resource disputes and make the environment a reason for cooperation rather than conflict.

Team travels to Indonesia to prep for judicial education course

Legal authorities are now available in Indonesia to enable civil society and the government to file claims to hold responsible parties liable for damages and the restoration of natural resources.

Through an ELI workshop and curriculum developed in conjunction with the Indonesian Center for Environmental Law and others, judges will learn best practices and methods for implementing new legal processes, including environmental damage valuation and restoration and compensation, tailored to the specific needs of the host country.

The goal is to promote environmental accountability through judicial enforcement. Ultimately, the benefits will include reduced deforestation and greenhouse gas emissions, as well as improved biodiversity and quality of life for vulnerable communities.

ELI recently traveled to Indonesia to help prepare for the week-long workshop to be held this summer. Staff met with various local stakeholders to gain background on topics like injury quantification, restoration and compensation, and settlement. ELI was also able to hear from judges which topics are most important to cover.

ELI staff held focus groups with ICEL as well as the Ministry of Environment and Forestry and Ministry of Justice and Human Rights, using an oil spill case to discuss valuation, settlement, and transboundary issues.

ELI and ICEL also held focus group discussions with the Supreme Court of Indonesia’s Environmental Working Group and Center for Training and Legal Research. The discussion included a presentation on the needs assessment by ICEL and a presentation on the comparative study of valuation, compensation, and restoration practice in several countries.

ELI’s judicial education program is a hallmark of the Institute’s work. With in-depth consultations, custom design of programs to meet the specific needs of the particular jurisdiction, and success in creating institutional capacity, the lessons learned continue to be applied after the education is completed. Since 1991, ELI has developed, presented, and participated in more than 40 workshops on critical topics in environmental law for more than 2,000 judges from 27 countries.

ELI met with a local NGO and members of the government to prepare for workshop on judicial enforcement of environmental laws.

Field Notes: Water summit showcases ELI legal expertise

ELI President Scott Fulton and Director of ELI’s Judicial Education Program Alejandra Rabasa traveled to Brazil to participate in the World Water Forum. The forum is the world’s biggest water-related event and is organized by the World Water Council, an international organization that brings together all those interested in the theme of water. Supreme Court justices from over 50 counties were in attendance to shine a light on the importance of rule of law in advancing water quality goals.

ELI hosted a day-long conference on Environmental Law In Practice in Detroit. The conference presented a spectrum of emerging legal issues with a focus on environmental justice. It introduced a wide-ranging exploration of career opportunities in the EJ field. This event featured environmental law experts on panels including Careers in Environmental Justice, Energy & Climate Justice, Water Access and Affordability, and Urban Air Quality.

Agustin V. Arbulu, executive director of the Michigan Department of Civil Rights, delivered opening remarks. Keynote addresses were given by Mustafa Santiago Ali, senior vice president of climate, environmental justice and community revitalization, Hip Hop Caucus, and Charles Lee, senior policy advisor, EPA Office of Environmental Justice.

Members of the public came together with lawyers, students, academics, civil rights and social justice advocates and activists, and community groups to discuss pressing issues.

The Conference was co-sponsored by Wayne State University Law School’s Transnational Environmental Law Clinic and Environmental Law Society, University of Chicago Law School’s Abrams Environmental Law Clinic, the American Bar Association’s Environmental Justice Committee of the Section of Civil Rights and Social Justice, and the Great Lakes Environmental Law Center.

Director of the Ocean Program Xiao Recio-Blanco moderated a webinar on Current Developments on U.S. Fisheries Policy. The Trump administration’s approach to fisheries management seems to constitute a significant policymaking shift. Recent decisions such as extending the Gulf of Mexico season for red snapper or overturning a decision by the Atlantic States Marine Fisheries Commission that would have cut New Jersey’s recreational quota for summer flounder seem to go against NOAA’s traditional approach of situating scientific information at the center of fisheries decisionmaking.

The webinar discussed these and other recent developments and assessed the direction U.S. fisheries policymaking may take in the future.

ELI and the China Environmental Protection Foundation held the first training session to build the capacity of public interest groups and prosecutors in China since receiving its temporary registration for an environmental protection-related project from China’s Ministry of Environmental Protection and the Beijing Bureau of Public Security.

The session was held at Tianjin University Law School. A total of 53 participants — comprising representatives from public interest groups, environmental courts, prosecutors, and environmental protection bureaus — attended from 16 provinces, autonomous regions, and cities.

Report on perils, promise of artificial intelligence.

Of Econs and Humans — Theory, Moral Licensing, and Crowding Out
Author
Bruce Rich
Current Issue
Issue
1
Bruce Rich

The award of the 2017 Nobel Prize for economics to Richard Thaler highlights the recognition that human behavior is more complex and irrational than the assumptions of conventional, still prevalent economic models. Psychologist Daniel Kahneman, who received the economics Nobel in 2002, described his astonishment that “our two disciplines seemed to be studying different species, which . . . Richard Thaler later dubbed Econs and Humans.” Kahneman warned of “theory induced blindness: once you accept a theory it is extraordinarily difficult to notice its flaws.”

This debate is of extraordinary importance for the future of various national and international environmental programs. Many may rely on assumptions that are questionable, and in some cases perverse in their environmental consequences.

A 2013 study in Energy Policy examined the impacts of a water conservation program in a Massachusetts housing complex on overall electricity consumption, including impacts on household CO2 emissions. Participants reduced their water consumption by 6 percent, but electricity use increased by 5.6 percent compared with a control group. The reduced water use cut energy consumption for hot water heating, but overall increases in electricity use exceeded these energy reductions by two-fold. Taking into account EPA estimates of electricity lost in production, transmission, and delivery, energy use increases with associated upstream losses were six times greater than the energy saved through reduced hot water heating.

The authors state that “adoption of a more environment friendly choice in one domain may actually increase the likelihood of less environment friendly behavior in other areas.” At root is “moral licensing,” whereby people feel that socially or morally praiseworthy efforts in one area lessens the need to act in other areas. They warn that “a considerable amount” of environmental programs and funds might “actually have a much smaller — or even a negative impact on CO2 emissions” than evaluations indicate.

As early as 1970, Robert M. Titmus, a professor at the London School of Economics, concluded that economic incentives for blood donors in the UK adversely affected, i.e., “crowded out,” donors’ main motivation: altruism and civic duty. Increasing payments for blood donations, he found, can actually decrease blood supply. Since then evidence has grown that some economic-incentive programs to promote publicly desirable activities are perverse in their consequences.

Studies in Switzerland, Wisconsin, Nevada, and Washington state on public approval of proposed nuclear waste disposal facilities found higher acceptance when the motivation is the greater public good or civic duty, and that the introduction of economic incentives reduces public acceptance, sometimes more than 50 percent.

In the past, many economists examined empirically expressed preferences without analyzing the impact of external material incentives on intrinsic psychological and cultural values. Yet this distinction, and the interrelationship between external and internal motivation, is critical for understanding how a program to change environmental behavior may work. Already in 1997, a Swiss study concluded that “where public spirit prevails,” using price incentives may crowd out civic duty, and actually make achieving the goals of a particular project more difficult. The authors urged that price incentives should be “reconsidered in all areas where intrinsic motivation can empirically shown to be important.”

Sometimes it is argued that while the crowding out phenomenon might apply to populations in richer countries, developing nations would be more responsive to economic incentives. A 2015 American Political Science Review article concluded that private economic incentives in a World Bank eco-development project in the Himalayas had a “negative and significant” impact on existing non-economic psychological and cultural motivation to conserve forests in the project population.

Those reporting that the project reduced formerly intrinsic cultural motivations for forest conservation in favor of economic incentives had “lower levels of conservationist behavior” than those who continued to cite the primacy of cultural and environmental values. Collective and community material benefits did have a positive correlation on environmental motivation. But community approaches to resource management, as opposed to private ownership models, are often not favored by governments and aid agencies.

The Indian and American authors warn that many sustainable- and ecodevelopment projects appear to be based on unfounded, simplistic assumptions about the correlation between small household economic incentives and achieving environmental goals. In fact, such “sustainable development projects can have the perverse effect of undermining their own environmental goals.”

Of econs and humans — theory, moral licensing, and crowding out.