Theoretical physicist Stephen Hawking and Tesla Motors founder Elon Musk have endorsed a number of principles that should ensure that artificial intelligence (AI) does not harm humans and only serves the good. The Silicon Valley Business Journal reports.
This pact is necessary in order to ensure the safety of people, as well as to answer many questions regarding AI. Musk is confident that AI can potentially be more dangerous than nuclear weapons. Hawking also holds a similar opinion and claims that AI can destroy humanity. However, both recognize that AI will help humans solve many problems - for example, cure cancer or slow down global warming.
Adherence to all 23 principles should ensure that AI only serves good purposes. They are broken down into three categories: Research Questions, Ethics and Values Questions, and Long Term Issues. The first category includes answers to the questions facing the creators of AI, including how the funding should proceed. The second category touches on security issues and ways to neutralize the threat. The third describes the possible risks that humanity may face when AI reaches its peak and the so-called superintelligence appears.
These principles were developed at a conference on AI discussion in early January 2017. DeepMind CEO Demis Hassabis contributed to the writing.
