Reading Group

The AISafety.com Reading Group meets weekly, usually Wednesdays at 19:45 UTC. To join, add “soeren.elverlin” on Skype.

Usually, we start with small-talk and a presentation round, then the host gives a summary of the paper for roughly half an hour. The summary of the article is uploaded on the Youtube Channel. This is followed by discussion (both on the article and in general) and finally we decide on a paper to read the following week.

Join us by Skype, by adding ‘soeren.elverlin’. Also check out our Facebook Group.

TITLE AUTHOR DATE SLIDES PRESENTATION
We’re told to fear robots. But why do we think they’ll turn on us? Steven Pinker 01-03-2018
Cognitive Biases Potentially Affecting Judgment of Global Risks Eliezer Yudkowsky 21-02-2018 https://www.dropbox.com/s/jrfcqwca4dxrvjq/Cognitive_biases.pptx?dl=0 https://youtu.be/z0B5xVdNq1E
Goodhart Taxonomy Scott Garrabrant 13-02-2018 https://www.dropbox.com/s/7bdeyrmrm8sccse/Goodhart%20Taxonomy.pptx?dl=0 https://youtu.be/tzbMSObyp4o
An AI Race for Strategic Advantage: Rhetoric and Risks Seán S ÓhÉigeartaigh et al. 07-02-2018 https://www.dropbox.com/s/vtxaak9ngppxyku/Rhetoric_and_Risks.pptx?dl=0 https://youtu.be/7wGdkoPxOSQ
Reply to Bostrom’s arguments for a hard takeoff Brian Tomasik 31-01-2018 https://www.dropbox.com/s/y8iltly2919zd3h/Reply_To_Bostrom.pptx?dl=0 https://youtu.be/04nu87UnslI
Superintelligence as a Cause or Cure for Risks of Astronomical Suffering Kaj Sotala et al. 24-01-2018 https://www.dropbox.com/s/1ekz8282oh207wy/Astronomical_Suffering.pptx?dl=0 https://youtu.be/Onx24F7Hliw
Impossibility of deducing preferences and rationality from human policy Stuart Armstrong et al. 17-01-2018 https://www.dropbox.com/s/3jhoj03vj8ph9du/2018-01-17-Stuart-Armstrong-You-cannot-learn-human-rationality-and-reward-together.pptx?dl=0 https://youtu.be/H7b_2NCJk1E
On the Promotion of Safe and Socially Beneficial Artificial Intelligence Seth Baum 09-01-2018 https://www.dropbox.com/s/c4b9cggo9ylc3jf/Beneficial_AI.pptx?dl=0 https://youtu.be/QYkvsEHxHxI
Refuting Bostrom’s Superintelligence Argument Sebastian Benthall 03-01-2018 https://www.dropbox.com/s/zcq5d0z6i66pu0j/Refuting%20Bostrom.pptx?dl=0 https://youtu.be/DfZjVKsnjTk
Logical Induction (1+7) Scott Garrabrant et al. 27-12-2017 https://www.dropbox.com/s/9jrm8whuswaytjb/Logical_Induction.pptx?dl=0 https://youtu.be/jbZyfcDcaKE
Conceptual Confusions in Assessing AGI Chris Cooper 20-12-2017 https://www.dropbox.com/s/ax49ckl3exhk9s0/Conceptual_Confusions.pptx?dl=0 https://youtu.be/d4luYznO1ec
Disjunctive Scenarios of Catastrophic AI Risk (2/2) Kaj Sotala 06-12-2017 https://www.dropbox.com/s/vaxifnfhqx5c3qk/Disjunctive_Scenarios2.pptx?dl=0 https://youtu.be/PcT6L5YKN9A
Disjunctive Scenarios of Catastrophic AI Risk (1/2) Kaj Sotala 01-12-2017 https://www.dropbox.com/s/wyw9ke0beapqglp/Disjunctive_Scenarios.ppsx?dl=0 https://youtu.be/T85G-ah3ynE
Artificial Intelligence in Life Extension: from Deep Learning to Superintelligence Alexey Turchin 22-11-2017 https://www.dropbox.com/s/p2e6ehi8gz27pk3/AI_in_Healthcare.pdf?dl=0 https://youtu.be/ZRy7ET-0KPI
Good and safe uses of AI Oracles Stuart Armstrong 08-11-2017 https://www.dropbox.com/s/xnl2t89z4nfbexr/Oracles.pdf?dl=0 https://youtu.be/5R73YhldO_o
Positively shaping the development of artificial intelligence Robert Wilbin 01-11-2017 https://www.dropbox.com/s/jk23r5xbmcrs37n/Positively_Shaping.pdf?dl=0 https://youtu.be/j_YyAv0CUG8
There is no Fire Alarm for Artificial General Intelligence Eliezer Yudkowsky 25-10-2017 https://www.dropbox.com/s/oqwmybfjq5tjebp/No_Fire_Alarm.pdf?dl=0 https://youtu.be/k86VzPd0wTQ
Fitting Values to Inconsistent Humans Stuart Armstrong 18-10-2017 https://www.dropbox.com/s/u3n9s9nsp009z0p/Fitting_Values.pdf?dl=0 https://youtu.be/w0H6Ws7xts8
Age of Em (Intelligence Explosion) Robin Hanson 11-10-2017 https://www.dropbox.com/s/o7se1q6996mc0ph/Age_of_Em_Intelligence_Explosion.pdf?dl=0 https://youtu.be/qlH-gyvZL1g
Age of Em (Chapter 27) Robin Hanson 27-09-2017 https://www.dropbox.com/s/i50zb7iswpivbgm/Age%20of%20Em.pdf?dl=0 https://youtu.be/vufsTEVK7DM
Meditations on Moloch Scott Alexander 20-09-2017 https://www.dropbox.com/s/nn0ep22szo2h9gs/Meditations_on_Moloch.pdf?dl=0 https://youtu.be/YQ_2HFtXBDM
Incorrigibility in the CIRL Framework Ryan Carey 13-09-2017 https://www.dropbox.com/s/yyryrngcs7qhsz5/Incorrigibility_In_CIRL.pdf?dl=0 https://youtu.be/n2X1QKEUrt4
OpenAI Makes Humanity Less Safe Ben Hoffman 06-09-2017 https://www.dropbox.com/s/v8ugc4uo5ds533b/OpenAI_Makes_Humanity_Less_Safe.pdf?dl=0 https://youtu.be/nwh9ZR3yO2M
Open Problems Regarding Counterfactuals: An Introduction For Beginners Alex Appel 30-08-2017 https://www.dropbox.com/s/0ztr9lwd9z8md2l/Counterfactuals.pdf?dl=0 https://youtu.be/JqtJXr9QHkM
A Game-Theoretic Analysis of the Off-Switch Game Tobias Wängberg et al. 23-08-2017 https://www.dropbox.com/s/n4rx49hq49m5bxi/Off-Switch-Game.pdf?dl=0 https://youtu.be/8w2_cb6cBY0
Benevolent Artificial Anti-Natalism Thomas Metzinger 16-08-2017 https://www.dropbox.com/s/6gfbt568tadnflo/Benevolent_Artificial_Anti-Natalism.pdf?dl=0 https://youtu.be/Zjid5CgLaac
Where the Falling Einstein Meets the Rising Mouse Scott Alexander 09-08-2017 https://www.dropbox.com/s/wu4pc3qc8zi13h0/Where_the_Falling_Einstein.pdf?dl=0  https://youtu.be/oua1fMxYXvo
Superintelligence Risk Project Jeff Kaufman 03-08-2017 https://www.dropbox.com/s/9jk00oohc912izx/Superintelligence_Risk_Project.pdf?dl=0 https://youtu.be/vG8SuD66NLA
Staring into the Singularity Eliezer Yudkowsky 26-07-2017 https://www.dropbox.com/s/f7g9mpcwk3qr9pk/Staring_Into_the_Singularity.pdf?dl=0 https://youtu.be/qud4WvehRho
Artificial Intelligence and the Future of Defense Matthijs Maas et al. 19-07-2017 https://www.dropbox.com/s/gxigsjyyd1thnb1/AI_and_the_Future_of_Defense.pdf?dl=0 https://youtu.be/UO6Px7-AL4w
Prosaic AI Alignment Paul Christiano 12-07-2017 https://www.dropbox.com/s/vlg3pb0pewx1w7r/Prosaic_AI_Alignment.pdf?dl=0 https://youtu.be/YvBj620UPBg
A model of the Machine Intelligence Research Institute Sindy Li 05-07-2017 https://www.dropbox.com/s/d2fxbnqyay104df/A_Model_of_MIRI.pdf?dl=0 https://youtu.be/kNlU3kAB2ks
Deep Reinforcement Learning from Human Preferences Paul Christiano et al. 28-06-2017 https://www.dropbox.com/s/ajajtd7fs3fhw8u/Deep_Reinforcement_Learning.pdf?dl=0 https://youtu.be/3zK1kNremWA
–Holiday– 21-06-2017
The Singularity: A Philosophical Analysis (2/2) David J. Chalmers 14-06-2017 https://www.dropbox.com/s/pan93bzvfroj58k/The_Singularity_2.pdf?dl=0
The Singularity: A Philosophical Analysis (1/2) David J. Chalmers 07-06-2017 https://www.dropbox.com/s/lu4qk2205htlku2/The_Singularity.pdf?dl=0 https://youtu.be/U-0ZD9Irfw
Why Tool AIs want to be Agent AIs Gwern Branwen 31-05-2017 https://www.dropbox.com/s/i9jvrj43r7xvocl/Tool_AIs.pdf?dl=0 https://youtu.be/Tnnn6LtZGiQ
A Map: AGI Failure Modes and Levels Alexey Turchin 24-05-2017 https://www.dropbox.com/s/to6qowvhh14wfut/AGI_Failure_modes.pdf?dl=0 https://youtu.be/kBTNrprdKiU
Neuralink and the Brain’s Magical Future Tim Urban 17-05-2017 https://www.dropbox.com/s/e00gsu629zkzl4b/Neuralink.pdf?dl=0 https://youtu.be/9NpNzlCptJI
The Myth of Superhuman AI Kevin Kelly 10-05-2017 https://www.dropbox.com/s/00cnhpyndlo4jru/The_Myth_of_a_Superhuman_AI.pdf?dl=0 https://youtu.be/WLSOmVXweSs
Merging our brains with machines won’t stop the rise of the robots Michael Milford 03-05-2017 https://www.dropbox.com/s/og3pn5o7ofi101e/Humans_Merging_with_AI.pdf?dl=0 https://youtu.be/Rgm6xMt54VA
Building Safe AI Andrew Trask 26-04-2017 https://www.dropbox.com/s/3fnx251f9oiga8p/Building_Safe_AI.pdf?dl=0 https://youtu.be/Ys-U-4vjRjw
AGI Safety Solutions Map Alexey Turchin 19-04-2017 https://www.dropbox.com/s/ldyb7a32nd2089k/AGI_Safety_Solutions_Map.pdf?dl=0 https://youtu.be/ZNSfUiXZwz0
Strong AI Isn’t Here Yet Sarah Constantin 12-04-2017 https://www.dropbox.com/s/297amvxrl58wgil/Strong_AI_Isnt_Here_Yet.pdf?dl=0 https://youtu.be/GpuQlJ3IHBM
Robotics: Ethics of artificial intelligence Stuart Russell et al. 05-04-2017 https://www.dropbox.com/s/8t5o990d1hf7ew6/Robotics_Ethics_of_artificial_intelligence.pdf?dl=0 https://youtu.be/z_WhxqCWJ4s
Using machine learning to address AI risk Jessica Taylor 29-03-2017 https://www.dropbox.com/s/52k4u10f95c6fvb/Using_Machine_Learning.pdf?dl=0 https://youtu.be/vXNi4L5PH0A
Racing to the Precipice: a Model of Artificial Intelligence Development Armstrong et al. 22-03-2017 https://www.dropbox.com/s/2zybpfb667vy9tl/Racing_To_The_Precipice.pdf?dl=0
Politics is Upstream of AI Raymond Brannen 15-03-2017 https://www.dropbox.com/s/kvcyf4kwmqmlufx/Politics_Is_Upstreams_of_AI.pdf?dl=0
Coherent Extrapolated Volition Eliezer Yudkowsky 08-03-2017 https://www.dropbox.com/s/2jldifzkpc82rmk/Coherent_Extrapolated_Volition.pdf?dl=0
–Cancelled due to illness– 01-03-2017
Towards Interactive Inverse Reinforcement Learning Armstrong, Leike 22-02-2017 https://www.dropbox.com/s/ouom3qzx8aofulv/Towards_Interactive_Inverse_Reinforcement_Learning_.pdf?dl=0
Notes from the Asilomar Conference on Beneficial AI Scott Alexander 15-02-2017 https://www.dropbox.com/s/4ohpo4fpewwdz7q/Notes_from_the_Asilomar_Conference_on_Beneficial_AI.pdf?dl=0
My current take on the Paul-MIRI disagreement on alignability of messy AI Jessica Taylor 08-02-2017 https://www.dropbox.com/s/9jtu8njaloxucrv/My_Current_take_on_the_Paul_MIRI_disagreement.pdf?dl=0
How feasible is the rapid development of Artificial Superintelligence? Kaj Sotala 01-02-2017 https://www.dropbox.com/s/5u79rex6czszt23/How_Feasible_is_the_Rapid_Development_of_Artificial_Superintelligence.pdf?dl=0
Response to Cegłowski on superintelligence Matthew Graves 25-01-2017 https://www.dropbox.com/s/bzlw8mc7k1fs0ox/Response_to_Ceglowski.pdf?dl=0
Disjunctive AI scenarios: Individual or collective takeoff? Kaj Sotala 18-01-2017 https://www.dropbox.com/s/sdsm2mpaiq892o3/Individual_or_collective_takeoff.pdf?dl=0
Policy Desiderata in the Development of Machine Superintelligence Nick Bostrom 11-01-2017 https://www.dropbox.com/s/jt6w0fzli5b0vg1/Policy%20Desiderata.pdf?dl=0
Concrete Problems in AI Safety Dario Amodei et al. 04-01-2017 https://www.dropbox.com/s/wthme4pnhlipz2q/Concrete.pdf?dl=0
–Holiday– 28-12-2016
A Wager on the Turing Test: Why I Think I Will Win Ray Kurzweil 21-12-2016 https://www.dropbox.com/s/iurbqzyaq9tt69f/Kurzweil.pdf?dl=0
Responses to Catastrophic AGI Risk: A Survey Sotala, Yampolskiy 14-12-2016 https://www.dropbox.com/s/iywy8znxx8yn1xt/Responses%20to%20AI.pdf?dl=0
Discussion of ‘Superintelligence: Paths, Dangers, Strategies’ Neil Lawrence 07-12-2016 https://www.dropbox.com/s/pyhb55mz65bhe9m/Neil%20Lawrence%20-%20Future%20of%20AI.pdf?dl=0
Davis on AI capability and motivation Rob Bensinger 30-11-2016 https://www.dropbox.com/s/eatjziiqsj5bmmg/Rob%20Bensinger%20Reply%20to%20Ernest%20Davis.pdf?dl=0
Ethical guidelines for a Superintelligence Ernest Davis 22-11-2016 https://www.dropbox.com/s/7j14li21igzi5gx/Ethical%20Guidelines%20for%20a%20Superintelligence.pdf?dl=0
Superintelligence: Chapter 15 Nick Bostrom 15-11-2016 https://www.dropbox.com/s/5jsusue656rdf2r/15%20Crunch%20Time.pdf?dl=0
Superintelligence: Chapter 14 Nick Bostrom 09-11-2016 https://www.dropbox.com/s/l2myz5c7t3a6at9/14%20Science%20and%20Technology%20Strategy.pdf?dl=0
Superintelligence: Chapter 11 Nick Bostrom 01-11-2016 https://www.dropbox.com/s/vj9j5saz39ese5i/11%20Multipolar%20Scenarios.pdf?dl=0
Superintelligence: Chapter 9 (2/2) Nick Bostrom 25-10-2016 https://www.dropbox.com/s/ux66z2ujz9jgofe/9.%20Motivation%20Selection%20Methods.pdf?dl=0
Superintelligence: Chapter 9 (1/2) Nick Bostrom 18-10-2016 https://www.dropbox.com/s/0mgnqcq075vehfv/Capability%20Control%20Methods.pdf?dl=0
Superintelligence: Chapter 8 Nick Bostrom 11-10-2016 https://www.dropbox.com/s/ihj35vxbevfghal/Default%20doom.pdf?dl=0
Superintelligence: Chapter 7 Nick Bostrom 04-10-2016 https://www.dropbox.com/s/pps6di0pza7wvab/The%20superintelligent%20Will.pdf?dl=0
Superintelligence: Chapter 6 Nick Bostrom 27-09-2016
Superintelligence: Chapter 5 Nick Bostrom 20-09-2016
Taxonomy of Pathways to Dangerous Artificial Intelligence Roman V. Yampolskiy 13-09-2016
Unethical Research: How to Create a Malevolent Artificial Intelligence Roman V. Yampolskiy 06-09-2016
Superintelligence: Chapter 4 Nick Bostrom 30-08-2016
Superintelligence: Chapter 3 Nick Bostrom 23-08-2016
Superintelligence: Chapter 1+2 Nick Bostrom 16-08-2016
Why I am skeptical of risks from AI Alexander Kruel 09-08-2016
–Break due to family extension– 02-08-2016
–Break due to family extension– 26-07-2016
Intelligence Explosion FAQ Luke Muehlhauser 19-07-2016
A toy model of the treacherous turn Stuart Armstrong 12-07-2016
The Fable of the Dragon Tyrant Nick Bostrom 05-07-2016
The Fun Theory Sequence Eliezer Yudkowsky 28-06-2016
Intelligence Explosion Microeconomics Eliezer Yudkowsky 21-06-2016
Strategic Implications of Openness in AI Development Nick Bostrom 14-06-2016
That Alien Message Eliezer Yudkowsky 07-06-2016
The Value Learning Problem Nate Soares 31-05-2016
Decisive Strategic Advantage without a Hard Takeoff Kaj Sotala 24-05-2016