23 principles to ‘best manage AI in coming decades’

MANILA, Philippines – Recently, an artificial intelligence (AI) computer program called Libratus defeated 4 poker professionals – a “landmark step” for AI, said its creator, because poker had been particularly challenging for AI up until that point.

AI’s victory in the poker contest – following wins in chess and the even more complex boardgame “go” – is a reminder that AI is getting smarter all the time. How then does humanity make sure that it remains an ally and doesn’t go rogue like Skynet did in the Terminator series? 

That was the question discussed by global experts at the Beneficial AI conference held in early January in California. The conference, in its second iteration, sought to hear out experts from diverse backgrounds to come up with a set of principles that will guide the development of AI. The video embedded above features one of the discussions held during the conference, in which experts discuss potential outcomes when AI has reached human-level capabilities. 

The conference was organized by the Future of Life (FOL) Institute – a non-profit organization founded by Massachusetts Institute of Technology (MIT) astronomist Max Tegmark, Skype cofounder Jaan Tallinn, and DeepMind research scientist Viktoriya Krakovna in March 2014, according to Business Insider.

The institute, on its website, described the process in distilling principles for AI: “We gathered all the reports we could and compiled a list of scores of opinions about what society should do to best manage AI in coming decades. From this list, we looked for overlaps and simplifications, attempting to distill as much as we could into a core set of principles that expressed some level of consensus.”

A principle is only included in the final list – called the “23 Asilomar AI Principles” because the conference was held at the Asilomar conference venue in California – if it gains the approval of 90% of the conference participants. 

The principles have since been endorsed by nearly 2,300 people, said Gizmodo, including 880 robotics and AI researchers. FOL scientific advisory board members, physicist Stephen Hawking and SpaceX CEO Elon Musk, have also endorsed the principles, listed below: 

Research Issues

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:

• 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?

3. Science-Policy Link: There should be constructive and healthy exchange between AI researchers and policy-makers.

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 Values

6. 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.

Longer-term Issues 

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. 

The principles are currently just that – principles. At this point, there is no governing body that will seek the enforcement of the proposed guidelines. Nevertheless, these principles can serve as a backbone or a starting point for potential laws or rules pertaining to the development of AI in the future.

“We hope that these principles will provide material for vigorous discussion and also aspirational goals for how the power of AI can be used to improve everyone’s lives in coming years,” the FOL’s website said. – Rappler.com