5 min readUsing AI for Social Good

On July 7, 2016, the Computing Community Consortium (CCC) held a workshop entitled Artificial Intelligence for Social Good. In attendance were over 300 participants, and an additional 3,500 audiences via live streaming. Results of the workshop, linked below this article, was released in March 2017.

On July 7, 2016, the Computing Community Consortium (CCC) held a workshop entitled Artificial Intelligence for Social Good. In attendance were over 300 participants, and an additional 3,500 audiences via live streaming. Results of the workshop, linked below this article, was released in March 2017.

The CCC hopes to highlight the societal benefits of AI, as well as provide guidance to future research and policy decisions involving the use of AI. According to the workshop, there are four essential public concerns that can benefit the most with AI deployment: urban computing, health, environmental sustainability, and public welfare. This report highlights the opportunities and challenges related to AI, as well as the cross cutting issues that need to be addressed in order to fully benefit from the such technological use.

AI in Urban Computing

Urban computing refers to the “study and application of computing technology in urban areas”. In the 2016 workshop, the discussion on urban computing centered on transport networks, and how it can be used to improve mobility and safety.

urban computng
AI can help improve flow and mobility in big cities

Some of the technologies related to urban computing are:

  • data analytics referring to travel patterns
  • pilot systems to optimize flow of traffic in cities
  • multi-modal transportation systems.

The group on urban computing stresses that there are various technical challenges that need to be addressed before AI can be fully deployed in this sector. For one, transportation systems are complex, and they are affected by human behavior. It is necessary to create models of urban computing that incorporates various transportation models, such as pedestrians, bicycles, cars, vans, and buses in order to determine how each could affect mobility in big cities.

AI in Sustainability Planning

The term “sustainability” often refers to the conservation of an ecosystem, but for this workshop, it was defined broadly “to include all aspects of sustainable biological, economic, and social systems that support human well being”. The session on sustainability focused mainly on the ecological aspect of sustainability.

Through AI, planners can utilize huge amounts of data found in government databases to create models to test out proposed interventions. This way, policies are streamlined and optimized, with unintended consequences appropriately considered.

Some of the challenges of AI deployment in this sector include:

  • development of methods that can collect and model data for a wider range of species, and integration of data from various sources
  • the need for a “systems of systems approach which incorporates social, cultural, and economic activity
  • biases in the quality of data, particularly those which come from crowd sourcing
  • lack of technical infrastructure such as poor networking, and the lack of skilled personnel with appropriate training and education in handling data
  • creation of a business model that support long-term data collection and policy enforcement efforts.

AI in Health

AI technologies has been used in the health care industry for years, but its application is largely commercial. As AI in health care improves, the government must up-scale deployment of such technologies to benefit the greater public. Some of the important opportunities for AI in the health care sector include:

  • targeted therapy options
  • new sensors and new health care delivery
  • personalized health care
  • collaborative and evidence-based decision-making

There are of course several concerns when it comes to AI in health service delivery. There is a need to address the biases in data, privacy protection, as well as cross-disciplinary re-training of medical practitioners. Perhaps the biggest challenge of all is the need for a learning health care system where a person’s data can be objectively analyzed, the right interventions and care are  dispatched, and feedback from the intervention can be re-captured, and future decisions adjusted based on feedback.

AI in Public Welfare

social good
In many countries, a large percentage of the population need the help of the government to provide for their own basic needs.

AI can help address various social issues such as justice, economic development, workforce development, public safety, policing, and education. But there are also real gaps that must be addressed in order for AI to be effective at addressing  such issues. Some of these include the lack of experienced collaborators, visible activity and case studies, reusable infrastructure, longitudinal perspective, and regulatory hurdles.

Why It Matters

We do not discount the fact that artificial intelligence will have tremendous impact on the society. This report from the CCC summarizes the many important benefits AI can have to the public and the provision of social goods. The question is: is social good the primary agenda of AI research? Much of the problems in the use of AI mentioned in this article requires dialogue of various governmental organizations and their stakeholders, but so far, no such dialogue is taking place.

Governments are busy creating AI weapons, and once again, public welfare takes a back seat. This is the greatest of our contentions. With all the wonderful possibilities  and the current concerns in the use of AI for social good, why are autonomous weapons a priority?

In order to truly assess how AI can be utilized to better the lives of the public we must ask:

  • What is your government’s main goal for AI development?
  • What AI technologies are they developing?
  • How are the issues of AI deployment mentioned in this article being addressed?

Leave a Reply

Your email address will not be published. Required fields are marked *