At the Beneficial AI Conference recently, Elon Musk, Stuart Russell, Ray Kurzweil, Demis Hassabis, Sam Harris, Nick Bostrom, David Chalmers, Bart Selman, and Jaan Tallinn chatted with moderator Max Tegmark what likely outcomes might be if we succeed in building human-level AGI, and also what we would like to happen.
Last month a panel of experts gathered at the Beneficial AI Conference, in Asilomar, California organized by the Future of Life Institute to discuss the most important issue of this century. The amazing group of AI researchers from academia and industry, and thought leaders in economics, law, ethics, and philosophy for five days dedicated to beneficial AI.
Below, Elon Musk, Stuart Russell, Ray Kurzweil, Demis Hassabis, Sam Harris, Nick Bostrom, David Chalmers, Bart Selman, and Jaan Tallinn discuss with moderator Max Tegmark what likely outcomes might be if we succeed in building human-level AGI, and also what we would like to happen.
When Tegmark polls the panel about when this may happen, the consensus (apart from Musk who plays to the crowd) is that this will happen in the time frame of years. Tegmark comments that the timescale makes a huge difference—hard takeoff versus soft takeoff.
"We're talking about human AI. Human AI is by definition at human levels, therefore is human."Kurzweil talked about how the growth of AGI might be better in a slow takeoff scenario. "As technologists we should do everything we can to keep the technology safe and beneficial. As we do each specific application, like self driving cars, there's a whole host of ethical issues to address, but I don't think we can solve the problem just technologically." Kurzweil projects that even if the most perfect and safe AI is created, it might be at the expense of the political system or other factors, it won't be an ideal outcome.
"We're talking about human AI. Human AI is by definition at human levels, therefore is human," he states. According to the futurist, the issue of how we make humans ethical is the same issue as how we make AIs human-level ethical.
In conjunction with the AI conference in Asilomar a large group of the leaders in AI and related fields teamed up and extended the open letter into a set of 23 principles for AI research, design and use, intended to ensure that AI lives up to its great potential to help and empower people in the decades and centuries ahead.
Artificial intelligence has already provided beneficial tools that we use every day by people around the world. According to the conference participants, continued development, guided by the following principles, will offer amazing opportunities to help and empower people in the decades and centuries ahead.
The Asilomar Principles
1) Research Goal: The goal of AI research should be to create not undirected intelligence, but beneficial intelligence.
2) Research Funding: Investments in AI should be accompanied by funding for research on ensuring its beneficial use, including thorny questions in computer science, economics, law, ethics, and social studies, such as:
3) Science-Policy Link: There should be constructive and healthy exchange between AI researchers and policy-makers.
- How can we make future AI systems highly robust, so that they do what we want without malfunctioning or getting hacked?
- How can we grow our prosperity through automation while maintaining people’s resources and purpose?
- How can we update our legal systems to be more fair and efficient, to keep pace with AI, and to manage the risks associated with AI?
- What set of values should AI be aligned with, and what legal and ethical status should it have?
4) Research Culture: A culture of cooperation, trust, and transparency should be fostered among researchers and developers of AI.
5) Race Avoidance: Teams developing AI systems should actively cooperate to avoid corner-cutting on safety standards.
Ethics and Values6) Safety: AI systems should be safe and secure throughout their operational lifetime, and verifiably so where applicable and feasible.
7) Failure Transparency: If an AI system causes harm, it should be possible to ascertain why.
8) Judicial Transparency: Any involvement by an autonomous system in judicial decision-making should provide a satisfactory explanation auditable by a competent human authority.
9) Responsibility: Designers and builders of advanced AI systems are stakeholders in the moral implications of their use, misuse, and actions, with a responsibility and opportunity to shape those implications.
10) Value Alignment: Highly autonomous AI systems should be designed so that their goals and behaviors can be assured to align with human values throughout their operation.
11) Human Values: AI systems should be designed and operated so as to be compatible with ideals of human dignity, rights, freedoms, and cultural diversity.
12) Personal Privacy: People should have the right to access, manage and control the data they generate, given AI systems’ power to analyze and utilize that data.
13) Liberty and Privacy: The application of AI to personal data must not unreasonably curtail people’s real or perceived liberty.
14) Shared Benefit: AI technologies should benefit and empower as many people as possible.
15) Shared Prosperity: The economic prosperity created by AI should be shared broadly, to benefit all of humanity.
16) Human Control: Humans should choose how and whether to delegate decisions to AI systems, to accomplish human-chosen objectives.
17) Non-subversion: The power conferred by control of highly advanced AI systems should respect and improve, rather than subvert, the social and civic processes on which the health of society depends.
18) AI Arms Race: An arms race in lethal autonomous weapons should be avoided.
19) Capability Caution: There being no consensus, we should avoid strong assumptions regarding upper limits on future AI capabilities.
20) Importance: Advanced AI could represent a profound change in the history of life on Earth, and should be planned for and managed with commensurate care and resources.
21) Risks: Risks posed by AI systems, especially catastrophic or existential risks, must be subject to planning and mitigation efforts commensurate with their expected impact.
22) Recursive Self-Improvement: AI systems designed to recursively self-improve or self-replicate in a manner that could lead to rapidly increasing quality or quantity must be subject to strict safety and control measures.
23) Common Good: Superintelligence should only be developed in the service of widely shared ethical ideals, and for the benefit of all humanity rather than one state or organization.
If you'd like to join Demis Hassabis, Yann LeCun, Yoshua Bengio, Stuart Russell, Peter Norvig, Ray Kurzweil, Jeff Dean, Tom Gruber, Francesca Rossi, Bart Selman, Leslie Kaelbling, Guru Banavar and others as a signatory, you'll find the principles and a signature form here.