Many friends of MIRI will know that I have been a supporter of its mission since its founding, and have rendered my informal assistance, including a major role in arranging matching funds for the Institute’s 2007 Challenge Grant.
I am pleased to take a more direct role in fostering its success as President of MIRI. I have left my previous role as Founder and Chief Strategist at SirGroovy.com, a growing online music licensing firm, and have been assuming responsibility for the management of the Machine Intelligence Research Institute over the last few weeks.
Prospective volunteers, donors, and aspiring researchers should now make contact with me rather than with Tyler Emerson.
To those who I am greeting for the first time, let me introduce myself. On a professional level, I hold a Master of Business Administration from Drexel University, and am coming from a role that combined management, research, analysis, and strategy in a fast-growing music licensing firm from its founding in New York. In that capacity, in a previous role at Aon, and in my academic studies I have been enduringly interested in finance and economics, particular the economics of technology and IP. Scientifically, I earned my undergraduate degree in biochemistry and have worked in several labs, as well as serving at the National Institute of Standards and Technology, and have extensively studied the history of science and technology, as well as the potential for biological cognitive enhancement. I have served with the Peace Corps in Kazakhstan, and am splitting my time between Manhattan, where I live with my wife Aruna, and Silicon Valley.
My interest in the safety of technological development, driven by the potentially grave ethical consequences, is over a decade old. I have been particularly focused on the potential of advanced nanotechnology and artificial intelligence, participating in forums such as Transvision and Foresight conferences, the SL4 mailing list, Overcoming Bias, and organizations such as MIRI and the Center for Responsible Nanotechnology (CRN). I plan to make my relevant work available at a single site, but in the meantime I will point to a small selection. For instance, I coauthored an analysis of the risks of advanced molecular manufacturing and mitigating strategies with Robert Freitas , and contributed “Corporate Cornucopia” as a member of CRN’s Global Task Force. Those who would like to see more can view Michael Anissimov’s archive of some of my writings at his website, and some of my most recent talks, an Institute of Ethics and Emerging Technologies presentation on the political implications of different conceptions of willpower and a Convergence08 talk on decision theory for humans, are available on the web.
As President, I plan to build on MIRI’s successes, such as the Singularity Summit, while also working to increase its internal and extramural research capabilities and output. In the course of the latter, I shall pay particular attention to the publication of research that improves the quality of our thinking about the potential and safety of advanced artificial intelligence, such as MIRI Research Fellow Eliezer Yudkowsky’s two contributions to the Oxford edited volume, Global Catastrophic Risks, and thus to better communicating internal research progress to our supporters.
Some of this work will involve indirect, meta-level contributions. For instance, recent work at MIRI by Rolf Nelson, Anna Salamon, Steven Rayhawk, Thomas McCabe and others has led to the development of a software tool for combining judgments about particular future scenarios and technological developments to reveal inconsistencies and enable the adoption of a more coherent probability assignment for planning. The content and algorithms of this tool have been completed, and work is now underway to finalize the software’s interface and make it publicly available to improve the quality of reasoning about interrelated technology scenarios, including those involving artificial intelligence.
Another effort involves conducting expert elicitation research to determine the state of academic and non-academic expert opinion regarding timelines and risks for advanced artificial intelligence.
Future research along these lines may explore particular biases and psychological factors affecting attitudes and reasoning related to artificial intelligence.
Other research will be directly focused on object-level problems. I plan to work vigorously to identify more promising extramural scholars whose work can be fruitfully promoted by MIRI grants, work such as MIRI-Canada Academic Prize Recipient Shane Legg’s “Machine Super-Intelligence.” At the same time I will be working to recruit and make best use of talented Research Fellows for MIRI’s internal efforts.
I look forward to describing further directions over the coming months, and invite the advice and opinions of the friends of MIRI at firstname.lastname@example.org.