Advancing beneficial technology
Foresight Institute is a leading non-profit researching and fostering technologies of fundamental importance for the human future, focusing on molecular machine nanotechnology, cybersecurity, artificial intelligence.
If we push in the right directions — learning, teaching, arguing, shifting directions, and pushing further — then we may yet steer the technology race toward a future with room enough for our dreams.
—K. Eric Drexler Engines of Creation
Join our reflection on creating positive futures
News and Updates
The Trajectory of Civilization: Extinction, Race to the Bottom, or Upward Climbing? Thank you for this incredible discussion, especially to our panelists Robin Hanson, Paul Christiano, Peter Ecklesurname, Christine Peterson, Mark S. Miller, Alyssa Vance. ... See MoreSee Less
Nobel laureate Sir Fraser Stoddart comments on Foresight: we could not be prouder! <3
youtu.be/1Wng24ASteQNobel laureate Sir Fraser Stoddart on Foresight Institute and Foresight Workshps. ... See MoreSee Less
Can we evolve our humanity ethically? How should we do such thing?
Watch David Eagleman (NeoSensory, @The Brain - PBS) and Arvind K Gupta (IndieBio) probed by Allison Dman (Foresight Institute) reflect on the ethics of future neuro & biotech.
This is pt. 2 of Making the most of our humanity - Foresight Institute X IndieBio. Pt. 1 here: youtu.be/Mlficse9JlAPart 2 of Foresight Institute's salon "Making the most of our Humanity" Held at IndieBio on June 13th 2018 ... See MoreSee Less
Foresight Institute added 3 new photos.
Ever wondered what being human will look like in the decades to come? Here's a peek into the future of our biology:
Watch David Eagleman (NeoSensory, The Brain - PBS) and Arvind K Gupta (IndieBio) probed by Lou Viquerat (Foresight Institute) to lay out a roadmap for neuro & biotech.
This is pt. 1 of Making the most of our humanity - Foresight Institute X IndieBio. Pt. 2 here: youtu.be/1SY10lFpN40Part 1 of Foresight Institute's salon "Making the most of our Humanity" Held at IndieBio on June 13th 2018 ... See MoreSee Less
Allison's beautiful intro to our salon" Making the most of our Humanity".
Topics: X-hope - humanity and beyond - Neurotech - Biotechyoutube.com ... See MoreSee Less
David Eagleman, Arvind Gupta, Allison Duettmann and Lou Viquerat from Foresight Institute joined for an evening to explore the path forward for human biology.
The evening had two practical purposes: 1/ laying out a roadmap of biotech and neurotech in order to 2/ establish an ethical framework that will guide the development of those technologies in the very near future. All participants were encouraged to inform the discussion around building the ethical framework. ... See MoreSee Less
Foresight Institute is at Indie.Bio.
A wholehearted thank you to all who made yesterday's salon pure magic: to our attendees for coming and asking probing questions, to our guests David Eagleman and Arvind K Gupta for answering those and ours with such brio, to Maya Lockwood for making it happen and Alex Kopelyan for opening the festivities!
Lou & Allison ... See MoreSee Less
Thank you to the participants of our strategy meeting on AGI Coordination for making this meeting into what it was: A proficient venture into coordination with an excellent group. White paper coming soon.
Afanasjeva, Olga Good AI
Armstrong, Stuart Future of Humanity Institute
Baum, Seth Global Catastrophic Risk Institute
Belfield, Haydn Centre for the Study of Existential Risk
Bensinger, Rob Machine Intelligence Research Institute
Bourgon, Malo Machine Intelligence Research Institute
Bowerman, Niel 80,000 Hours
Braley, Ryan Lightbend
Brown, Tom Google Brain
Burja, Samo Bismarck Analysis
Carey, Ryan Ought, Foresight Fellow
Cooper, Betsy Center for Long-Term Cybersecurity
Cuperman, Miron Base Zero
Cussins, Jessica Future of Life Institute, Center for Long-Term Cybersecurity
Ding, Jeffrey Future of Humanity Institute
Duettmann, Allison Foresight Institute
Eckersley, Peter Electronic Frontier Foundation
Fischer, Kevin Crypto Lotus
Flynn, Carrick Future of Humanity Institute
Garfinkel, Benjamin Future of Humanity Institute
Guan, Melody Stanford
Irving, Geoffrey OpenAI
Kai, De Hong Kong University of Science & Technology
Kotran, Alex AI Initiative, Harvard Kennedy School
Krakovna, Victoria DeepMind
Kramar, Janos DeepMind
Lai, Tony Legal.io
Leung, Jade Future of Humanity Institute
Liston, Matthew ConsenSys
Maas, Matthijs Future of Humanity Institute, Global Catastrophic Risk Institute
Mallah, Richard Future of Life Institute
Mangan, Fiona Justice and Security in Transitions
McCaslin, Tegan AI Impacts
McReynolds, Joe China Security Studies Fellow, Jamestown Foundation
Michaud, Eric Rift Recon
Miller, Mark Foresight Institute, Agoric
Mosleh, Ali John Garrick Institute for the Risk Sciences
Nitzberg, Mark Center for Human Compatible AI
O'Neill, Jim Mithril Capital
Page, Michael OpenAI
Peterson, Christine Foresight Institute
Scheyer, Peter Foresight Fellow
Shulman, Carl Future of Humanity Institute
Singh, Tanya Future of Humanity Institute
Tallinn, Jaan Future of Life Institute, Center for the Study of Existential Risk
Vance, Alyssa Apprente
Webb, Michael Stanford
Xiao, Qiang School of Information, UC Berkeley, China Digital Times
Xu, Mimee UnifyID
Yampolskiy, Roman University of Louisville, Foresight Fellow ... See MoreSee Less
I really feel that the ' Strategy meeting on AGI Coordination' should look like this:
Designing an artificial nose that would replicate the functions of a dog's nose?
Just one of the fascinating things Fellow in molecular machines Chris Wilmer is building through his work on large scale computational screening of new molecular system.
Christopher Wilmer is currently an Assistant Professor in Chemical Engineering at the University of Pittsburgh where he leads the Hypothetical Materials Lab. He received his Ph.D. in Chemical & Biological Engineering from Northwestern University and BASc in Engineering Science from the University of Toronto. He was the co-founder of NuMat Technologies and co-founding editor of the journal Ledger. Since he first read Eric Drexler’s Engines of Creation and Ray Kurzweil’s The Age of Spiritual Machines, he wanted to become a physicist. He believes the world needs more engineers developing nanotechnology.Christopher Wilmer is currently an Assistant Professor in Chemical Engineering at the University of Pittsburgh where he leads the Hypothetical Materials Lab.... ... See MoreSee Less
Can we make clouds reflect sunlight to combat the adverse effects of Global Warming?
Proud to say that Berhane Temelso, Foresight Fellow Computational Chemistry is working on using aerosol particles to this effect, and not just that.
Berhane Temelso is currently a computational chemistry research scientist working at Furman University. He received his Ph.D. in chemistry from Georgia Institute of Technology and B.A. in physics from Berea College. His Ph.D. work explored the ability of the most rigorous first-principles computational methods to reproduce molecular properties derived from experiment. His current research is mainly focused on the application of efficient computational methods to understand the structure and dynamics of hydrogen-bonded systems ranging from water clusters to atmospheric aerosols. A large part of his efforts are devoted to understanding the effect of atmospheric aerosols on the global radiation balance and ways in which they can be used to engineer solutions to mitigate climate change. Berhane is an author on early papers on hydrogen abstraction and donation tools for the mechanosynthesis of diamondoids published in The Journal of Physical Chemistry A.Berhane Temelso is currently a computational chemistry research scientist working at Furman University. He received his Ph.D. in chemistry from Georgia Insti... ... See MoreSee Less
Proud to introduce our Fellow, Eva Maria Strauch:
Eva Maria works on understanding how a virus’s proteins encode information to hijack specific cells - a way to identify its Achilles heel to stop viral infections & learn new ways to fight cancer cells.
Eva-Maria Strauch is currently an acting instructor/translational investigator at the University of Washington’s Institute for Protein Design. She received her Ph.D. from the University of Texas at Austin’s Dept. of Chemistry and Biochemistry. She is inspired by the meticulous atomic details by which proteins are arranged and their involvement in almost all processes in life.Eva-Maria Strauch is currently an acting instructor/translational investigator at the University of Washington’s Institute for Protein Design. She received h... ... See MoreSee Less
We are pleased to invite you to our salon with David Eagleman, Arvind Gupta at IndieBio on June 13th. Follow this link for program, speakers and tickets: mailchi.mp/foresight/salonindiebio ... See MoreSee Less
What could solve at once money issues, diseases, the limitations of our physical bodies, and death?
See far and big into the future through the eyes of Michael Skuhersky, our visionary 2017 Foresight Fellow in Mind Uploading.
Michael Skuhersky is a graduate student in the Synthetic Neurobiology group at the MIT Media Lab. Previously, he was at the MIT Center for Bits and Atoms, working on protein docking dynamics, rapid fabrication techniques, and spatial computing architectures. Before that, he conducted research in various areas of particle physics at MIT and UCLA. He is passionate about devoting his time to making mind uploading a reality in his lifetime; currently he is attempting to build and validate an emulation of a simple brain that will respond to and deliver external stimuli in the exact same way as that of a real organism. He has also dabbled in practical applications of transhumanism while running a novel prosthetics startup aimed at the Chinese market. In his spare time, he goes hiking, studies world politics, and composes experimental music.Michael Skuhersky is a graduate student in the Synthetic Neurobiology group at the MIT Media Lab. Previously, he was at the MIT Center for Bits and Atoms, wo... ... See MoreSee Less
Foresight Institute added 3 new photos.
"I am just the detective that gets to trawl through the literature, and figure out: what precise molecular mechanisms can we target that would statistically save the most lives?"
So pleased to introduce you to Cosmo Mielke, 2017 Foresight Fellow in Longevity, a truly bright mind with an extraordinary agility working across scientific disciplines - something we admire and encourage here at Foresight Institute.
Cosmo Mielke is currently the founder of Infino.me, a nonprofit research project to study aging with user donated quantified self data. Cosmo received his PhD in Bioinformatics and Computational Biology from the Biodesign Institute and the Mayo Clinic, and a B.S. in Astronomy, Physics and Mathematics from the University of Arizona. He is also a data scientist in the UCSF Memory and Aging Center, where he manages the computational pipelines to analyze MRI scans of Alzheimer’s patients. He is inspired most by the rapidly changing landscape of science.Cosmo Mielke is currently the founder of Infino.me, a nonprofit research project to study aging with user donated quantified self data. Cosmo received his Ph... ... See MoreSee Less
We like to look at problems through a practical, solution oriented perspective
- it was thus a great pride, to count Kimberly Hamad-Schifferli as Foresight Fellow in Biomolecular Machines in our class of 2017.
Kimberly Hamad-Schifferli is currently Associate Professor of Engineering, College of Science at University of Massachusetts Boston and Visiting Scientist, Department of Mechanical Engineering, MIT. Her areas of expertise are nanotechnology, nanobiotechnology, nanomaterials, and rapid diagnostics. Kimberly received her Ph.D. in Chemistry from the University of California at Berkeley, and an S.B. in Chemistry from MIT. Her mission is to create a scientific legacy, which means training people that are not only skilled and creative, but also have a strong sense of citizenship and integrity.Kimberly Hamad-Schifferli is currently Associate Professor of Engineering, College of Science at University of Massachusetts Boston and Visiting Scientist, D... ... See MoreSee Less
"Feynman said that human beings, in the future, should be able to control at the atom level. Currently we are still far away from that."
Meet Chuyang Chen, 2017 Foresight Fellow in Molecular Machines, bringing Feynman's vision a step closer to existence.
Chuyang Cheng is currently a Postdoc Researcher working in Sir Fraser Stoddart’s group at Northwestern University, from which he received his Ph.D. in Chemistry. He received his BS in Chemistry from Peking University. His research focus is on design and synthesis molecular machines as well as incorporating molecular machines into functional materials. Mr. Cheng was the recipient of the 2015 Distinguished Student Award from the Foresight Institute. He is most enthusiastic about controlling molecules to do what he expects them to do precisely.Chuyang Cheng is currently a Postdoc Researcher working in Sir Fraser Stoddart’s group at Northwestern University, from which he received his Ph.D. in Chemis... ... See MoreSee Less
As our fellowship class of 2017 ends, here is a human we would like you to know:
Jonathan Barnes, 2017 Fellow in Synthetic Polymer Chemistry. It was an honor to count such a brilliant mind amongst our fellows, and a pleasure to close this fellowship by the presentation of the 2018 Feynman prize at the side of Nobel Laureate Sir Fraser Stoddart.
Jonathan Barnes is currently an Assistant Professor in the Department of Chemistry at Washington University in St. Louis. Jonathan received his Ph.D. in Organic Chemistry from Northwestern University, an MS in Organic Chemistry and B.S. in Chemistry both from the University of Kentucky. Inspired by the art of synthetic organic chemistry, Jonathan looks outside of the physical sciences as inspiration for making functional organic materials and molecular machines that may be used to solve problems in chemistry, energy, and human health.Jonathan Barnes is currently an Assistant Professor in the Department of Chemistry at Washington University in St. Louis. Jonathan received his Ph.D. in Orga... ... See MoreSee Less
Potential opening position!
3-month contract with possible full-time position later to support Foresight's projects; looking for someone who: strongly aligns with our mission, is well-connected in tech or tech-related philanthropy; science & tech background a big plus. Welcome to share but don't reply here; email to email@example.com ... See MoreSee Less
Poorly written ad, no details of what Mission to be "closely aligned with" and no job description link. BIG FAT FAIL..
There is no link to a job description / RFP.
If you're too stupid to click on the ad, and do a little research, you're absolutely not a candidate for the job. Cheers!
cc Rob Lach
Summary session of the seminar on Artificial General Intelligence & Corporations at the Internet Archive.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteSummary session of the seminar AGI & Corporations (foresight-agi-corporations.eventbrite.com) with: - Allison Duettmann - AI Safety Researcher at For... ... See MoreSee Less
Mark Miller and Allison Duettmann discuss civilization as relevant superintelligence, the importance of cybersecurity, and decentralized approaches to AI safety.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteMark S. Miller and Allison Duettmann discuss civilization as relevant superintelligence, the importance and neglectedness of cybersecurity, and decentralized... ... See MoreSee Less
Tom Kalil speaking at The Internet Archive on incentivizing governmental institutions into adopting a long term project mindset.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteTom Kalil at the Internet Archive on aligning long-term projects with incentives in governmental institutions. Watch the rest of our Seminar on AGI and Corpo... ... See MoreSee Less
Mark Nitzberg at the Internet Archive presenting CHAI's (Center for Human-Compatible AI) approach to the traditional field of AI Alignment.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteMark Nitzberg (CHAI) giving an overview of the traditional field of AI alignment. Seminar on Artificial General Intelligences & Corporations hosted by Foresi... ... See MoreSee Less
Peter Scheyer exploring frontiers between AI, AGI, superintelligence, and collective intelligence, aligning corporation's with human interests and AI Safety.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteCorporations as Artificial General Intelligences (based on a literature review for a grant given by Paul Christiano on the legal aspects of AGI as corporatio... ... See MoreSee Less
Elizabeth Enayati Powers at the Internet Archive giving an overview of corporate personhood & definitions.
Watch the rest of our Seminar on AGI and Corporations on our Youtube channel: www.youtube.com/c/ForesightInstituteElizabeth Enayati Powers at the Internet Archive giving an overview of Corporate Personhood & Definitions Seminar on Artificial General Intelligences & Corpo... ... See MoreSee Less
Allison Duettmann at the Internet Archive, exploring the jungle of AI definitions, and giving an overview of the field of AGI safety.
Watch the rest our seminar on Corporations and AGI on our Youtube channel: www.youtube.com/c/ForesightInstituteAllison Duettmann giving a quick overview: - of the jungle of AI definitions, including AI, AGI, Superhuman and Superintelligence and how inclusive the defin... ... See MoreSee Less
2018 Integrated Molecular Machines Workshop: An incredible weekend of ideation and action - from an Ask Me Anything with Sir Fraser Stoddart, to setting the goal of integrated molecular machines, to surveying the state-of-the-art research, to generating five top novel project proposals that push the field toward integration - thank you to everyone who participated. ... See MoreSee Less
Are details of the workshop available anywhere?
Joon Yun, Thomas Kalil, Melanie Swan, Matt Bell discuss the toolkit to make it through the century and beyond.
Topics covered: X-Risks and X-Hope - prioritization - meta-toolsJoon Yun - MD, President, Palo Alto Investors Thomas Kalil - Senior Advisor, Eric & Wendy Schmidt Group Melanie Swan - Founder, Institute for Blockchain Stud... ... See MoreSee Less
David Eagleman, Robert McIntyre, Randal Koene, Nell Watson on brains, other computers, and the intersection between them.
Topics covered: State of the art AI - self-directed evolution - AI safetyDavid Eagleman - Author, Neuroscientist, Stanford University Robert McIntyre - CEO, Nectome Randal Koene - Lead Scientist, Kernel Nell Watson - Co-Founder, O... ... See MoreSee Less
Zooko Wilcox, Nathana Sharma, Mark Miller, Nick Kritikos, Allison Duettmann on how to build better base realities block by block.
Topics covered: Beyond crypto - ICO hype & hope - decentralized governanceZooko Wilcox - Founder, ZCash Nathana Sharma - Program Director, Singularity University Mark Miller - Senior Fellow, Foresight Institute Nick Kritikos, Head ... ... See MoreSee Less
Sonia Arrison Senkut, Aubrey de Grey, Max More, Kevin Perrott compare strategies to extend human healthspans @Vision Weekend.Sonia Arrison - Author, Advisor at Singularity University, Thiel Foundation Aubrey De Grey, Co-Founder of SENS Research Foundation Max More, President & CEO ... ... See MoreSee Less
The 4 minutes summary of our 2017 Vision Weekend is here! We are so grateful for all who attended and contributed.
Full recordings of our panels on Longevity, Blockchain, Intelligence, and the Tools to Make it Beyond the Century are available on our YouTube channel:
www.youtube.com/playlist?list=PLH78wfbGI1x25QAIGcnNn9OZAP_CXKpP7 ... See MoreSee Less
The second in our Debunking X Salon Series, this time with Robin Hanson - on exposing our primate brains' hidden motives and social agendas.
youtu.be/iRDIhxeerBAMonday, March 19, 2018 in San Francisco "Our Primate Brain: Enlightening Hidden Motives & Social Agendas" a Debunking X Salon by Foresight Institute An inter... ... See MoreSee Less
Allison Dman's SXSW workshop is online, divided into three parts:
1. Overview of AI safety and why the field can be broken down into the four focus areas Ethics, Technical Alignment, Cybersecurity, and Social Coordination
2. Overview of state of the art research in the focus areas
3. An alternative approach to AI safety that proposes to decentralize the danger, brought forth by Mark S. Miller, Christine Peterson, and Allison Dman.
bit.ly/aiphilosophyThis workshop was held in March 2018 at SXSW in Austin and is divided into three parts: Part 1 starts with an overview of the main problems in the four main ... ... See MoreSee Less
Allison Dman today at Simulation. Covering philosophy, foresight, existential hope, AI safety, riffs of utopia, strengthening civilizations, longevity, meta-ethics, and the simulation. www.youtube.com/watch?v=UWPqnOG6nZM&feature=youtu.beAllison is a AI Safety Researcher at Foresight Institute. www.existentialhope.com/ ****** Simulation inspires YOU to build the future. For entreprene... ... See MoreSee Less
Allison Dman today at Simulation. Covering philosophy, foresight, existential hope, AI safety, riffs of utopia, strengthening civilizations, longevity, the blockchain ecosystem and the simulation. www.youtube.com/watch?v=O_S1dsvvLwIAllison is a AI Safety Researcher at Foresight Institute. www.existentialhope.com/ ... See MoreSee Less
TOMORROW: What makes us treat corporations as legal persons? Does this entail anything for our treatment of future Artificial General Intelligences? Elizabeth Powers will set the stage for this discussion at our seminar on Artificial General Intelligences & Corporations @Internet Archive. Some tickets are still left - we look forward to tackling this complex but important thematic with you. ... See MoreSee Less
75 talk: Elizabeth Powers: Fundamental definitions of corporate personhood: Artificial General Intelligences & Corporations at the Internet Archive: Sunday, April 8, 2018 1052 3:07 pm 1032 4/8/18 30 conference: Artificial General Intelligences & Corporations at the Internet Archive: Sunday, April 8, 2018 4 Elizabeth Powers business law 69 intellectual property 2 corporate law corporations are legal I don't get how they go together to give some background information to talk about corporations artificial general intelligence entities the different types of business entities that exist 6 business entities what's the correlation? sole proprietorships they need to pay taxes on that let's share liability 11 liability what are the obligations that business entities have generally? it's created by individuals a relationship that's created by individuals 3 a separate entity 2 as a separate entity existence as a separate entity it is a legal construct spreading out the liability 1 C corporations 1 S corporations to spread that liability out to spread that liability out in a different way that legal status natural persons legal entities a supreme court case 14 personhood the existence of a corporation 58 in the past certain constitutional rights privileges and immunities the Citizens United case 1 a wealthy individual a collection of individuals as a united front 1 a lot more compelling corporations don't vote corporations don't run for office participation in the political process 13 it depends it depends on what that environment is the context of liability the context of ethics it's a collection of individuals what the corporation does 5 the corporation 1 the shareholders 5 a board of directors the corporation as independent entity 2 shareholders 1 our political process an investment in a mutual fund the directors and officers of corporations 1 a fiduciary duty a duty of loyalty a duty of due care the business judgment rule those duties which you have some lever of accountability within the corporate structure within that legal entity payment to play regulatory schemes 1 a little bit of background 1 Elizabeth Enayati Powers 1 legal persons
How to reorient the general thrust of AI research towards provably beneficial systems?
Berkeley's Center for Human Compatible AI answer is that it implies refocusing AI away from the capability to achieve arbitrary objectives and towards the ability to generate provably beneficial behavior. 🤖❣️
Because the meaning of beneficial depends on properties of humans, this task inevitably includes elements from the social sciences in addition to AI. 👩💼🧠
It is thus an intricate problem, and we look forward to hearing Mark Nitzberg disentangle its components for us at the Artificial General Intelligences & Corporations @Internet Archive seminar next Sunday!🌞 ... See MoreSee Less
sound like a good event wish we had advance notice to attend to short of notice for me.
139 talk: Mark Nitzberg: Overview of the Traditional Field of AI Alignment: Artificial General Intelligences & Corporations at the Internet Archive: Sunday, April 8, 2018 1169 3:49 pm 1032 4/8/18 30 conference: Artificial General Intelligences & Corporations at the Internet Archive: Sunday, April 8, 2018 AI Alignment & CHAI 17 chai CHAI: center for human compatible artificial intelligence 54 Norbert Wiener a mechanical agency the purpose into the machine 1 AI alignment Mark Nitzberg: firstname.lastname@example.org the expected outcome utility that utility function 4 Stuart Russell very difficult to pin down its own continued existence the paperclip maximizer instrumental goals if you do the thought experiment 2 so what can we do? the orthogonality thesis if you give a system an objective final goals ways in which we're making progress in AI safety Vingean reflection safe interruptibility self modification human-robot cooperation a sub-discipline to develop the conceptual and technical wherewithal... the concept of safety we are just getting started 1 Future of Life Institute long-term risks from AI doing technical research there's a lot of math on the whiteboard producing fundamental research 3 what we don't do some legislation that's pending the nearer-term AI issues research example cooperative inverse reinforcement learning (CIRL) algorithms for reinforcement learning inverse reinforcement learning CIRL: cooperative inverse reinforcement learning Ng and Russell (2000) defining a reward function defining a reward function that captures complex human values their reward function much more realistic 9 a payoff the off-switch game the off-switch problem 6 arxiv.org arxiv.org/abs/1611.08219 never completely certain if the human doesn't switch me off to prove mathematically utility maximizers how does this all link to corporations? ruthlessly maximizing its utility to bring corporations into alignment 11 from our perspective to reinvent the corporation 4 our research in orbit around us our invitation unsafe AI safe AI the incentives of the utility maximizer a rational utility maximizer there are always loopholes there are always loopholes that a rational utility maximizer will naturally learn to exploit 61 the power the power of an AGI to behave in a way we approve of 1 its incentive structure to regulate corporations the generalist if you write something in a journal to plead the fifth the values of humans and society the reason we are doing what we are doing to give more power to intelligent systems 2 to weigh in aligned systems 1 more powerful systems grounding in human values 48 human values 1 sort of a mixed bad an artificial general intelligence smart humans to require smart humans 1 the normative questions to bring into alignment in our interest retrospectively to retrospectively force the system to behave in a way we approve of trying to retrospectively force the system to behave in a way we approve of we are trying to retrospectively force the system to behave in a way we approve of if we're trying to reason with it a one-sentence question really long comments the way to think about what we're doing 3 as we delegate more and more decisions to intelligent systems the entirety of your values having the same result 1 Nitzberg 1 Center for Human Compatible AI Berkeley's Center for Human Compatible AI UC Berkeley's Center for Human Compatible AI 198 UC Berkeley 1 provably beneficial provably beneficial behavior to generate provably beneficial behavior the meaning of beneficial 2 provably beneficial systems 2 Executive Director of the UC Berkeley Center for Human Compatible AI 1 the UC Berkeley Center for Human Compatible AI UC Berkeley Center for Human Compatible AI human compatible AI the difference between directly using website to place an order and setting up an agent to place an order based on some criteria decision indirection levels of decision indirection level of decision indirection the criteria for making a decision the criteria for making an indirect decision the criteria for making a beneficial decision delegation of decision delegation of a decision delegation indirectness 62 Internet Archive 5 Sunday, April 8, 2018 2 April 8, 2018 arxiv.org/abs/1611.08219 paper: The Off-Switch Game paper: The Off-Switch Game: arxiv.org/abs/1611.08219 safer AI 2 cooperative inverse reinforcement learning
Creating counterculture - the first in our Debunking X Salon series, a discussion between Dr. Joon Yun and Allison Dman at The Embassy Network.
They discuss the evolutionary genealogy of our cultural paradigm, how technological progress resulted in current dislocated cultural equilibria, and how counterculture can give rise to a new social contract.
youtu.be/wFuATR56OwgAllison Duettmann, Foresight Institute in discussion with Joon Yun, MD, President of Palo Alto Investors - February 12th, 2018. In times of bitcoin frenzy, p... ... See MoreSee Less