Concluding Remarks

Last updated July 28, 2016

AI systems are augmenting human intelligence and will ultimately transform our personal and professional lives. Its benefits far outweigh its risks. And with the right policies and support, those benefits can be realized sooner.

Policy makers should focus on:

  • Facilitating a fact-based dialogue on the capabilities and limitations of AI technologies
  • Developing progressive social and economic policies to deploy AI systems for broad public good
  • Developing progressive education and workforce programs for future generations
  • Investing in a long-range interdisciplinary research program for advancing the science and design of AI systems

Our perspective is that AI enables a partnership between people and machines that boosts productivity, fosters new discoveries, and frees up resources that could be put to better use. While, in delivering on that promise, AI may displace jobs to other activities, it does not replace them. While, in pursuit of that partnership, AI certainly needs to be applied with care and thought, the risks that have been identified with it are often overstated and they can in any case be managed and mitigated.

Already, AI systems are part of our daily lives answering questions and making recommendations for products and services, and more are on the way to help people live and work “smarter” in a world where big data is the new natural resource.

Our approach to bringing AI systems to market involves pursuing a practical approach of building task-specific or domain-specific AI tools, focused on a range of industries and the professional tasks within them. In addition we are enabling a range of personal tasks for consumers (i.e., the customers of IBM’s enterprise clients). As we build these domain-specific AI systems, we are generalizing the underlying machinery and making it available on “horizontal” (cross-industry) platforms, in turn making it progressively easier to build a great variety of new domain-specific systems. From a technology perspective, building practical AI systems in the “cognitive” mold requires technologies that go well beyond textbook AI – in this paper, our perspective is informed by decades of investment in the underlying mathematical sciences; learning, reasoning and decision technologies; language, speech & vision technologies; human interface technologies; distributed and high-performance computing; and new computing architectures and devices.

This document has argued for policy makers to adopt a cognitive computing and cognitive systems approach as the main thrust to AI policy. Cognitive systems are technology tools that help people process large amounts of unstructured data to gain insights and make better decisions; helping people deal with complexity. We urge a people-centered systems redesign approach to Artificial Intelligence (AI) for Intelligence Augmentation (IA).

The progression from cognitive tool to assistant to collaborator to coach to mediator will happen gradually, since building cognitive systems is still very hard with many research questions, research gaps, data sets, and skillsets still missing. Nevertheless, the recommendations in this short document, if followed, will help us get the societal benefits of cognitive systems faster and safer.

Back to summary.