Deceptive Misaligned Mesa-Optimisers? Its More Likely Than You Think... Robert Miles AI Safety 2021-05-23 | The previous video explained why it's *possible* for trained models to end up with the wrong goals, even when we specify the goals perfectly. This video explains why it's *likely*.Previous video: The OTHER AI Alignment Problem: youtu.be/bJLcIBixGj8The Paper: arxiv.org/pdf/1906.01820.pdfMedia Sources:End of Ze World - youtu.be/enRzYWcVyAQFlexClip News graphicsWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesTimothy LillicrapKieryn James Scott WorleyJames E. PettsChad JonesShevis JohnsonJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidPeter RolfTeague LasserAndrew BlackledgeFrank MarsmanBrad BrookshireCam MacFarlaneCraig MederiosJon WrightCaptObviousJason HisePhil MoyerErik de BruijnAlec JohnsonClemens ArbesserLudwig SchubertAllen FaureEric JamesMatheson BayleyQeith Wreidjugettje dutchkingOwen Campbell-MooreAtzin Espino-MurnaneJohnny VaughanJacob Van BurenJonatan RIngvi GautssonMichael GreveTom O'ConnorLaura OldsJon HallidayPaul HobbsJeroen De DauwLupuleasa IonuțCooper LawtonTim NeilsonEric ScammellIgor KellerBen Glantonanul kumar sinhaTorDuncan OrrWill GlynnTyler HerrmannIan MunroJoshua DavisJérôme BeaulieuNathan FishPeter HozákTaras BobrovytskyJeremy Vaskó RichárdBenjamin WatkinAndrew HarcourtLuc RitchieNicholas GuyettJames Hinchcliffe12tone Oliver HabrykaChris BeachamZachary GidwitzNikita KiriyAndrew SchreiberSteve TrambertMario LoisBraden TisdaleAbigail NovickСергей УваровBela RMinkChris RimmerEdmund FokschanerGrant ParksJNate GardnerJohn AslanidesMaraErikBlnDragonSheep Richard NewcombeDavid MorganFionn Dmitri AfanasjevMarcel WardAndrew WeirKabsMiłosz WierzbickiTendayi MawusheJake FishWr4thon Martin OttosenRobert HildebrandtAndy KobreKees Darko SperacRobert ValdimarssonMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrik Daniel MunterAlex KnauthKasper Ian ReyesJames FowkesTom SayerLen Alan BandurkaBen HSimon PilkingtonDaniel KokotajloDiagon Andreas BlomqvistBertalan BodorZannheim Daniel Eickhardtlyon549 14zRobot Ivan Jason CherryIgor (Kerogi) Kostenkoib_ Thomas DingemanseStuart AlldrittAlexander BrownDevon BernardTed StokesJames HelmsJesper AnderssonDeepFriedJif Chris DinantRaphaël LévyJohannes WalterMatt StantonGarrett MaringAnthony ChiuGhaith TarawnehJulian SchulzStellated HexahedronCaleb Scott ViteriClay UptonConor ComiconorMichael RoeschterGeorg GrassIsak Matthias HölzlJim RenneyEdison FranklinPiers CalderwoodMikhail TikhomirovRichard OttoMatt BrauerJaeson BookerMateusz KrzaczekArtem HoncharovMichael WaltersTomasz GlinieckiMihaly BaraszMark WoodwardRanzearNeil PalmereRajeen NabidChristian EppleClark SchaeferOlivier CoutuIestyn bleasdale-shepherdMojoExMachinaMarek BelskiLuke PetersonEric EldardEric RogstadEric CarlsonCaleb LarsonMax ChiswickAronDavid de KloetSam FreedoslindenauA21 Johannes LindmarkNicholas TurnerTero KValerio GalieniFJannisM IRyan W AmmonsLudwig KrinnerThis person's name is too hard to pronouncekp contalloomlegsEverardo González ÁvalosKnut LøklingholmAndrew McKnightAndrei TrifonovAleks DMutual Informationpatreon.com/robertskmiles
Learn AI Safety at MATS #shorts Robert Miles AI Safety 2024-09-27 | Apply for MATS at matsprogram.org by Oct 6 #shorts
AI Ruined My Year Robert Miles AI Safety 2024-06-01 | How to Help: aisafety.info/questions/8TJV/How-can-I-helpaisafety.comAI Safety Talks: youtube.com/@aisafetytalksThere's No Rule That Says We'll Make It: youtube.com/watch?v=JD_iA7imAPsThe other "Killer Robot Arms Race" Elon Musk should worry about: youtube.com/watch?v=7FCEiCnHcboRob's Reading List:Podcast: rmrlp.libsyn.comYouTube Channel: youtube.com/@RobMilesReadingListThe FLI Open Letter: youtube.com/watch?v=3GHjhG6Vo40Yudkowsky in TIME: youtube.com/watch?v=a6m7JynBp-0Ian Hogarth in the FT: youtube.com/watch?v=Z8VvF82T6soLinks:The CAIS Open Letter: safe.ai/work/statement-on-ai-riskThe FLI Open Letter: futureoflife.org/open-letter/pause-giant-ai-experimentsThe Bletchley Declaration: gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration/the-bletchley-declaration-by-countries-attending-the-ai-safety-summit-1-2-november-2023US Executive Order: whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligenceSome analysis of the EO: thezvi.substack.com/p/on-the-executive-order"Sparks of AGI" Paper: arxiv.org/abs/2303.12712Yudkowsky in TIME: time.com/6266923/ai-eliezer-yudkowsky-open-letter-not-enoughHogarth in the FT: ft.com/content/03895dc4-a3b7-481e-95cc-336a524f2ac2The AI Safety Institute: gov.uk/government/publications/ai-safety-institute-overview/introducing-the-ai-safety-instituteResponsible Scaling Policies: metr.org/blog/2023-09-26-rspThe EU AI Act: https://artificialintelligenceact.eu/the-act/Hinton on CBS: youtu.be/qpoRO378qRYSources:"Sparks of AGI" Talk: youtube.com/watch?v=qbIk7-JPB2cYann LeCunn on Lex Fridman's Podcast: youtube.com/watch?v=SGzMElJ11CcWhite House Press Briefings: https://x.com/TVNewsNow/status/1663640562363252742youtube.com/watch?v=JHNkyHl5FpYKing Chuck on AI: youtube.com/watch?v=0_jw40Ga_mA"Equally sharing a cake between three people - Numberphile": youtube.com/watch?v=kaMKInkV7VsCommunity, various screenshotsThe SimpsonsSneakers (1992)Thanks to Rational Animations for the train sequence!youtube.com/@RationalAnimationsWith enormous thanks to my wonderful patrons:- Tor Barstad- Timothy Lillicrap- Juan Benet- Sarah Howell- Kieryn - Mazianni- Scott Worley- Jason Hise- Clemens Arbesser- Francisco Tolmasky- David Reid- Andrew Blackledge- Cam MacFarlane- Olivier Coutu- CaptObvious- Ze Shen Chin- ikke89- Isaac- Erik de Bruijn- Jeroen De Dauw- Ludwig Schubert- Eric James- Owen Campbell-Moore- Raf Jakubanis- Esa Koskinen- Nathan Metzger- Jonatan R- Gunnar- Laura Olds- Paul Hobbs- Bastiaan Cnossen- Eric Scammell- Alexare- Reslav Hollós- Jérôme Beaulieu- Nathan Fish- Taras Bobrovytsky- Jeremy - Vaskó Richárd- Andrew Harcourt- Chris Beacham- Zachary Gidwitz- Art Code Outdoors- Abigail Novick- Edmund Fokschaner- DragonSheep - Richard Newcombe- Joshua Michel- Richard - ttw- Sophia Michelle Andren- Alan J. Etchings- James Vera- Stumbleboots- Peter Lillian- Grimrukh- Colin Ricardo- DN- Mr Cats- Robert Paul Schwin- Roland G. McIntosh- Benjamin Mock- Emiliano Hodges- Maxim Kuzmich- Joanny Raby- Tom Miller- Eran Glicksman- CheeseBerry- Hoyskedotte- Alexey Malafeev- Jeff Starr- Justin - Liviu Macovei- Javier Soto- David Christal- Jam- Just Me- Sebastian Zimmer- Matt Thompson- Xan Atkinson- Andy- Albert Higgins- Alexander230- Clay Upton- Alex Ander- Carolyn- Nathan Rogowski- David Morgan- little Bang- Chad M Jones- Dmitri Afanasjev- Christian Oehne- Marcel Ward- Andrew Weir- Miłosz Wierzbicki- Tendayi Mawushe- Kees - loopuleasa- Marco Tiraboschi- Fraser Cain- Patrick Henderson- Daniel Munter- Ian- James Fowkes- Len - Yuchong Li- Diagon - Puffjanga- Daniel Eickhardt- 14zRobot - Stuart Alldritt- DeepFriedJif - Garrett Maring- Stellated Hexahedron- Jim Renney- Edison Franklin- Piers Calderwood- Matt Brauer- Mihaly Barasz- Rajeen Nabid- Iestyn bleasdale-shepherd- Marek Belski- Luke Peterson- Eric Rogstad- Max Chiswick- slindenau- Nicholas Turner- Jannis Funk- This person's name is too hard to pronounce- Jon Wright- Andrei Trifonov- Bren Ehnebuske- Martin Frassek- Matthew Shinkle- Robby Gottesman- Ohelig- Sarah - Nikola Tasev- Tapio Kortesaari- Soroush Pour- Boris Badinoff- DangerCat - Jack Phelps- Kyle Green- Lexi X- John Slape- Joel Gardner- Christopher Creutzig- Johann Puzik- Pindex - RMR- Andrew Edstrompatreon.com/robertskmiles
Apply to Study AI Safety Now! #shorts Robert Miles AI Safety 2023-04-28 | Apply to SERI MATS at serimats.org by May 7th, and check out http://aisafety.training to stay up to date with events and programs!
Why Does AI Lie, and What Can We Do About It? Robert Miles AI Safety 2022-12-09 | How do we make sure language models tell the truth?The new channel!: youtube.com/@aisafetytalksEvan Hubinger's Talk: https:/youtu.be/OUifSs28G30ACX Blog Post: astralcodexten.substack.com/p/elk-and-the-problem-of-truthful-aiWith thanks to my wonderful Patrons at http://patreon.com/robertskmiles :- Tor Barstad- Kieryn - AxisAngles - Juan Benet- Scott Worley- Chad M Jones- Jason Hise- Shevis Johnson- JJ Hepburn- Pedro A Ortega- Clemens Arbesser- Chris Canal- Jake Ehrlich- Kellen lask- Francisco Tolmasky- Michael Andregg- David Reid- Teague Lasser- Andrew Blackledge- Brad Brookshire- Cam MacFarlane- Olivier Coutu- CaptObvious- Girish Sastry- Ze Shen Chin- Phil Moyer- Erik de Bruijn- Jeroen De Dauw- Ludwig Schubert- Eric James- Atzin Espino-Murnane- Jaeson Booker- Raf Jakubanis- Jonatan R- Ingvi Gautsson- Jake Fish- Tom O'Connor- Laura Olds- Paul Hobbs- Cooper- Eric Scammell- Ben Glanton- Duncan Orr- Nicholas Kees Dupuis- Will Glynn- Tyler Herrmann- Reslav Hollós- Jérôme Beaulieu- Nathan Fish- Peter Hozák- Taras Bobrovytsky- Jeremy - Vaskó Richárd- Report Techies- Andrew Harcourt- Nicholas Guyett- 12tone - Oliver Habryka- Chris Beacham- Zachary Gidwitz- Nikita Kiriy- Art Code Outdoors- Andrew Schreiber- Abigail Novick- Chris Rimmer- Edmund Fokschaner- April Clark- John Aslanides- DragonSheep - Richard Newcombe- Joshua Michel- Quabl- Richard - Neel Nanda- ttw- Sophia Michelle Andren- Trevor Breen- Alan J. Etchings- Jenan Wise- Jonathan Moregård- James Vera- Chris Mathwin- David Shaffer- Jason Gardner- Devin Turner- Andy Southgate- Lorthock The Banisher- Peter Lillian- Jacob Valero- Christopher Nguyen- Kodera Software- Grimrukh- MichaelB- David Morgan- little Bang- Dmitri Afanasjev- Marcel Ward- Andrew Weir- Ammar Mousali- Miłosz Wierzbicki- Tendayi Mawushe- Wr4thon - Martin Ottosen- Alec Johnson- Kees - Darko Sperac- Robert Valdimarsson- Marco Tiraboschi- Michael Kuhinica- Fraser Cain- Patrick Henderson- Daniel Munter- And last but not least- Ian Reyes- James Fowkes- Len - Alan Bandurka- Daniel Kokotajlo- Yuchong Li- Diagon - Andreas Blomqvist- Qwijibo (James)- Zannheim - Daniel Eickhardt- lyon549 - 14zRobot - Ivan - Jason Cherry- Igor (Kerogi) Kostenko- Stuart Alldritt- Alexander Brown- Ted Stokes- DeepFriedJif - Chris Dinant- Johannes Walter- Garrett Maring- Anthony Chiu- Ghaith Tarawneh- Julian Schulz- Stellated Hexahedron- Caleb - Georg Grass- Jim Renney- Edison Franklin- Jacob Van Buren- Piers Calderwood- Matt Brauer- Mihaly Barasz- Mark Woodward- Ranzear- Rajeen Nabid- Iestyn bleasdale-shepherd- MojoExMachina- Marek Belski- Luke Peterson- Eric Rogstad- Caleb Larson- Max Chiswick- Sam Freedo- slindenau- Nicholas Turner- FJannis- Grant Parks- This person's name is too hard to pronounce- Jon Wright- Everardo González Ávalos- Knut- Andrew McKnight- Andrei Trifonov- Tim D- Bren Ehnebuske- Martin Frassek- Valentin Mocanu- Matthew Shinkle- Robby Gottesman- Ohelig- Slobodan Mišković- Sarah - Nikola Tasev- Voltaic- Sam Ringer- Tapio Kortesaarihttp://patreon.com/robertskmiles
Apply Now for a Paid Residency on Interpretability #short Robert Miles AI Safety 2022-11-11 | redwoodresearch.org/remixReal main channel video coming soon!#short #shorts
$100,000 for Tasks Where Bigger AIs Do Worse Than Smaller Ones #short Robert Miles AI Safety 2022-10-14 | Check out http://inversescaling.com for more details and to apply#shorts #short
Free ML Bootcamp for Alignment #shorts Robert Miles AI Safety 2022-05-24 | Apply for the second MLAB (Machine Learning for Alignment Bootcamp)!forum.effectivealtruism.org/posts/vvocfhQ7bcBR4FLBx/apply-to-the-second-iteration-of-the-ml-for-alignment
Win $50k for Solving a Single AI Problem? #Shorts Robert Miles AI Safety 2022-02-08 | The Alignment Research Center is offering up to $50k for proposals for their Eliciting Latent Knowledge problem.The report: docs.google.com/document/d/1WwsnJQstPq91_Yh-Ch2XRL8H_EpsnjrC1dwZXR37PC8/editContest details: alignmentforum.org/posts/QEYWkRoCn4fZxXQAY/prizes-for-elk-proposals#Shorts
Apply to AI Safety Camp! #shorts Robert Miles AI Safety 2021-11-19 | Trying out #shortsApplications are open for next year's AI Safety Camp!http://aisafety.camp
We Were Right! Real Inner Misalignment Robert Miles AI Safety 2021-10-10 | Researchers ran real versions of the thought experiments in the 'Mesa-Optimisers' videos!What they found won't shock you (if you've been paying attention)Previous videos on the subject:The OTHER AI Alignment Problem: Mesa-Optimizers and Inner Alignment: youtu.be/bJLcIBixGj8Deceptive Misaligned Mesa-Optimisers? It's More Likely Than You Think...: youtu.be/IeWljQw3UgQThe Paper: arxiv.org/abs/2105.14111The Interpretability Article: https://distill.pub/2020/understanding-rl-vision/Jacob Hilton's thoughts about what's going on: alignmentforum.org/posts/iJDmL7HJtN5CYKReM/empirical-observations-of-objective-robustness-failures?commentId=8f5B3skqocxJjK4mkAI Safety Camp: https://aisafety.camp/With thanks to my wonderful Patrons at http://patreon.com/robertskmiles :- Gladamas- Timothy Lillicrap- Kieryn - AxisAngles - James - Jake Fish- Scott Worley- James Kirkland- James E. Petts- Chad Jones- Shevis Johnson- JJ Hepboin- Pedro A Ortega- Clemens Arbesser- Said Polat- Chris Canal- Jake Ehrlich- Kellen lask- Francisco Tolmasky- Michael Andregg- David Reid- Peter Rolf- Teague Lasser- Andrew Blackledge- Brad Brookshire- Cam MacFarlane- Craig Mederios- Jon Wright- CaptObvious- Brian Lonergan- Girish Sastry- Jason Hise- Phil Moyer- Erik de Bruijn- Alec Johnson- Ludwig Schubert- Eric James- Matheson Bayley- Qeith Wreid- jugettje dutchking- James Hinchcliffe- Atzin Espino-Murnane- Carsten Milkau- Jacob Van Buren- Jonatan R- Ingvi Gautsson- Michael Greve- Tom O'Connor- Laura Olds- Jon Halliday- Paul Hobbs- Jeroen De Dauw- Cooper Lawton- Tim Neilson- Eric Scammell- Igor Keller- Ben Glanton- Tor Barstad- Duncan Orr- Will Glynn- Tyler Herrmann- Ian Munro- Jérôme Beaulieu- Nathan Fish- Peter Hozák- Taras Bobrovytsky- Jeremy - Vaskó Richárd- Benjamin Watkin- Andrew Harcourt- Luc Ritchie- Nicholas Guyett- 12tone - Oliver Habryka- Chris Beacham- Nikita Kiriy- Andrew Schreiber- Steve Trambert- Braden Tisdale- Abigail Novick- Serge Var- Mink- Chris Rimmer- Edmund Fokschaner- April Clark- J- Nate Gardner- John Aslanides- Mara- ErikBln- DragonSheep - Richard Newcombe- Joshua Michel- P - Alex Doroff- BlankProgram- Richard - David Morgan- Fionn - Dmitri Afanasjev- Marcel Ward- Andrew Weir- Kabs- Ammar Mousali- Miłosz Wierzbicki- Tendayi Mawushe- Wr4thon - Martin Ottosen- Andy K- Kees - Darko Sperac- Robert Valdimarsson- Marco Tiraboschi- Michael Kuhinica- Fraser Cain- Robin Scharf- Klemen Slavic- Patrick Henderson- Hendrik - Daniel Munter- Alex Knauth- Kasper - Ian Reyes- James Fowkes- Tom Sayer- Len - Alan Bandurka- Ben H- Simon Pilkington- Daniel Kokotajlo- Yuchong Li- Diagon - Andreas Blomqvist- Iras- Qwijibo (James)- Zubin Madon- Zannheim - Daniel Eickhardt- lyon549 - 14zRobot - Ivan - Jason Cherry- Igor (Kerogi) Kostenko- ib_ - Thomas Dingemanse- Stuart Alldritt- Alexander Brown- Devon Bernard- Ted Stokes- Jesper Andersson- DeepFriedJif - Chris Dinant- Raphaël Lévy- Johannes Walter- Matt Stanton- Garrett Maring- Anthony Chiu- Ghaith Tarawneh- Julian Schulz- Stellated Hexahedron- Caleb - Clay Upton- Conor Comiconor- Michael Roeschter- Georg Grass- Isak Renström- Matthias Hölzl- Jim Renney- Edison Franklin- Piers Calderwood- Mikhail Tikhomirov- Matt Brauer- Mateusz Krzaczek- Artem Honcharov- Tomasz Gliniecki- Mihaly Barasz- Mark Woodward- Ranzear- Neil Palmere- Rajeen Nabid- Clark Schaefer- Olivier Coutu- Iestyn bleasdale-shepherd- MojoExMachina- Marek Belski- Luke Peterson- Eric Rogstad- Eric Carlson- Caleb Larson- Max Chiswick- Aron- Sam Freedo- slindenau- Johannes Lindmark- Nicholas Turner- Intensifier- Valerio Galieni- FJannis- Grant Parks- Ryan W Ammons- This person's name is too hard to pronounce- contalloomlegs- Everardo González Ávalos- Knut Løklingholm- Andrew McKnight- Andrei Trifonov- Aleks D- Mutual Information- Tim - A Socialist Hobgoblin- Bren Ehnebuske- Martin Frassek- Sven Drebitz- Quabl- Valentin Mocanu- Philip Crawford- Matthew Shinkle- Robby Gottesman- Juanchihttp://patreon.com/robertskmiles
Intro to AI Safety, Remastered Robert Miles AI Safety 2021-06-24 | An introduction to AI Safety, remastered from a talk I gave at "AI and Politics" in LondonThe second channel: youtube.com/channel/UC4qH2AHly_RSRze1bUqSSNwExperts' Predictions about the Future of AI: http://youtu.be/HOJ1NVtlnyQ9 Examples of Specification Gaming: http://youtu.be/nKJlF-olKmgpatreon.com/robertskmilesWith thanks to my wonderful Patreon supporters:GladamasTimothy LillicrapKieryn AxisAngles James Nestor PoliticsScott WorleyJames KirklandJames E. PettsChad JonesShevis JohnsonJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidPeter RolfTeague LasserAndrew BlackledgeFrank MarsmanBrad BrookshireCam MacFarlaneCraig MederiosJon WrightCaptObviousBrian LonerganJason HisePhil MoyerErik de BruijnAlec JohnsonClemens ArbesserLudwig SchubertEric JamesMatheson BayleyQeith Wreidjugettje dutchkingOwen Campbell-MooreAtzin Espino-MurnaneJohnny VaughanCarsten MilkauJacob Van BurenJonatan RIngvi GautssonMichael GreveTom O'ConnorLaura OldsJon HallidayPaul HobbsJeroen De DauwCooper LawtonTim NeilsonEric ScammellIgor KellerBen GlantonTor BarstadDuncan OrrWill GlynnTyler HerrmannIan MunroJoshua DavisJérôme BeaulieuNathan FishPeter HozákTaras BobrovytskyJeremy Vaskó RichárdBenjamin WatkinAndrew HarcourtLuc RitchieNicholas GuyettJames Hinchcliffe12tone Oliver HabrykaChris BeachamZachary GidwitzNikita KiriyAndrew SchreiberSteve TrambertBraden TisdaleAbigail NovickSerge VarMinkChris RimmerEdmund FokschanerJNate GardnerJohn AslanidesMaraErikBlnDragonSheep Richard NewcombeJoshua MichelAlex AltairP David MorganFionn Dmitri AfanasjevMarcel WardAndrew WeirKabsAmmar MousaliMiłosz WierzbickiTendayi MawusheJake FishWr4thon Martin OttosenRobert HildebrandtAndy KobreKees Darko SperacRobert ValdimarssonloopuleasaMarco TiraboschiMichael KuhinicaFraser CainKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrik Daniel MunterAlex KnauthKasper Ian ReyesJames FowkesTom SayerLen Alan BandurkaBen HSimon PilkingtonDaniel KokotajloYuchong LiDiagon Andreas BlomqvistBertalan BodorQwijibo (James)Zubin MadonZannheim Daniel Eickhardtlyon549 14zRobot Ivan Jason CherryIgor (Kerogi) Kostenkoib_ Thomas DingemanseStuart AlldrittAlexander BrownDevon BernardTed StokesJesper AnderssonDeepFriedJif Chris DinantRaphaël LévyJohannes WalterMatt StantonGarrett MaringAnthony ChiuGhaith TarawnehJulian SchulzStellated HexahedronCaleb Scott ViteriClay UptonConor ComiconorMichael RoeschterGeorg GrassIsak RenströmMatthias HölzlJim RenneyEdison FranklinPiers CalderwoodMikhail TikhomirovMatt BrauerJaeson BookerMateusz KrzaczekArtem HoncharovMichael WaltersTomasz GlinieckiMihaly BaraszMark WoodwardRanzearNeil PalmereRajeen NabidChristian EppleClark SchaeferOlivier CoutuIestyn bleasdale-shepherdMojoExMachinaMarek BelskiLuke PetersonEric EldardEric RogstadEric CarlsonCaleb LarsonMax ChiswickAronSam FreedoslindenauA21 Johannes LindmarkNicholas TurnerIntensifierValerio GalieniFJannisGrant ParksRyan W AmmonsThis person's name is too hard to pronouncekp contalloomlegsEverardo González ÁvalosKnut LøklingholmAndrew McKnightAndrei TrifonovAleks DMutual InformationTim A Socialist HobgoblinBren EhnebuskeMartin FrassekSven Drebitzpatreon.com/robertskmiles
The OTHER AI Alignment Problem: Mesa-Optimizers and Inner Alignment Robert Miles AI Safety 2021-02-16 | This "Alignment" thing turns out to be even harder than we thought.# LinksThe Paper: arxiv.org/pdf/1906.01820.pdfDiscord Waiting List Sign-Up: https://forms.gle/YhYgjakwQ1Lzd4tJ8AI Safety Career Bottlenecks Survey: guidedtrack.com/programs/n8cydtu/run# Referenced VideosIntelligence and Stupidity - The Orthogonality Thesis: http://youtu.be/hEUO6pjwFOo9 Examples of Specification Gaming: youtu.be/nKJlF-olKmgWhy Would AI Want to do Bad Things? Instrumental Convergence: youtu.be/ZeecOKBus3QHill Climbing Algorithm & Artificial Intelligence - Computerphile: http://youtu.be/oSdPmxRCWwsAI Gridworlds - Computerphile: http://youtu.be/eElfR_BnL5kGenerative Adversarial Networks (GANs) - Computerphile: http://youtu.be/Sw9r8CL98N0# Other MediaThe Simpsons Season 5 Episode 19: "Sweet Seymour Skinner's Baadasssss Song"1970s Psychology study of imprinting in ducks. Behaviorism: http://youtu.be/2xd7o3z957cWith thanks to my excellent Patreon supporters:patreon.com/robertskmiles- Timothy Lillicrap- Gladamas- James - Scott Worley- Chad Jones- Shevis Johnson- JJ Hepboin- Pedro A Ortega- Said Polat- Chris Canal- Jake Ehrlich- Kellen lask- Francisco Tolmasky- Michael Andregg- David Reid- Peter Rolf- Teague Lasser- Andrew Blackledge- Frank Marsman- Brad Brookshire- Cam MacFarlane- Jason Hise- Phil Moyer- Erik de Bruijn- Alec Johnson- Clemens Arbesser- Ludwig Schubert- Allen Faure- Eric James- Matheson Bayley- Qeith Wreid- jugettje dutchking- Owen Campbell-Moore- Atzin Espino-Murnane- Johnny Vaughan- Jacob Van Buren- Jonatan R- Ingvi Gautsson- Michael Greve- Tom O'Connor- Laura Olds- Jon Halliday- Paul Hobbs- Jeroen De Dauw- Lupuleasa Ionuț- Cooper Lawton- Tim Neilson- Eric Scammell- Igor Keller- Ben Glanton- anul kumar sinha- Duncan Orr- Will Glynn- Tyler Herrmann- Tomas Sayder- Ian Munro- Joshua Davis- Jérôme Beaulieu- Nathan Fish- Taras Bobrovytsky- Jeremy - Vaskó Richárd- Benjamin Watkin- Sebastian Birjoveanu- Andrew Harcourt- Luc Ritchie- Nicholas Guyett- James Hinchcliffe- 12tone - Oliver Habryka- Chris Beacham- Zachary Gidwitz- Nikita Kiriy- Parker - Andrew Schreiber- Steve Trambert- Mario Lois- Abigail Novick- Сергей Уваров- Bela R- Mink- Fionn - Dmitri Afanasjev- Marcel Ward- Andrew Weir- Kabs- Miłosz Wierzbicki- Tendayi Mawushe- Jake Fish- Wr4thon - Martin Ottosen- Robert Hildebrandt- Poker Chen- Kees - Darko Sperac- Paul Moffat- Robert Valdimarsson- Marco Tiraboschi- Michael Kuhinica- Fraser Cain- Robin Scharf- Klemen Slavic- Patrick Henderson- Oct todo22- Melisa Kostrzewski- Hendrik - Daniel Munter- Alex Knauth- Kasper - Ian Reyes- James Fowkes- Tom Sayer- Len - Alan Bandurka- Ben H- Simon Pilkington- Daniel Kokotajlo- Peter Hozák- Diagon - Andreas Blomqvist- Bertalan Bodor- David Morgan- Zannheim - Daniel Eickhardt- lyon549 - Ihor Mukha- 14zRobot - Ivan - Jason Cherry- Igor (Kerogi) Kostenko- ib_ - Thomas Dingemanse- Stuart Alldritt- Alexander Brown- Devon Bernard- Ted Stokes- James Helms- Jesper Andersson- DeepFriedJif - Chris Dinant- Raphaël Lévy- Johannes Walter- Matt Stanton- Garrett Maring- Anthony Chiu- Ghaith Tarawneh- Julian Schulz- Stellated Hexahedron- Caleb - Scott Viteri- Conor Comiconor- Michael Roeschter- Georg Grass- Isak - Matthias Hölzl- Jim Renney- Edison Franklin- Piers Calderwood- Krzysztof Derecki- Mikhail Tikhomirov- Richard Otto- Matt Brauer- Jaeson Booker- Mateusz Krzaczek- Artem Honcharov- Michael Walters- Tomasz Gliniecki- Mihaly Barasz- Mark Woodward- Ranzear- Neil Palmere- Rajeen Nabid- Christian Epple- Clark Schaefer- Olivier Coutu- Iestyn bleasdale-shepherd- MojoExMachina- Marek Belski- Luke Peterson- Eric Eldard- Eric Rogstad- Eric Carlson- Caleb Larson- Braden Tisdale- Max Chiswick- Aron- David de Kloet- Sam Freedo- slindenau- A21 - Rodrigo Couto- Johannes Lindmark- Nicholas Turner- Tero Kpatreon.com/robertskmiles
Quantilizers: AI That Doesnt Try Too Hard Robert Miles AI Safety 2020-12-13 | How do you get an AI system that does better than a human could, without doing anything a human wouldn't?A follow-up to "Maximizers and Satisficers": youtu.be/Ao4jwLwT36MThe Paper: intelligence.org/files/QuantilizersSaferAlternative.pdfMore about this area of research: alignmentforum.org/tag/mild-optimizationWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesTimothy LillicrapGladamasJames Scott WorleyChad JonesShevis JohnsonJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidPeter RolfTeague LasserAndrew BlackledgeFrank MarsmanBrad BrookshireCam MacFarlaneVivek NayakJason HisePhil MoyerErik de BruijnAlec JohnsonClemens ArbesserLudwig SchubertAllen FaureEric JamesMatheson BayleyQeith Wreidjugettje dutchkingOwen Campbell-MooreAtzin Espino-MurnaneJohnny VaughanJacob Van BurenJonatan RIngvi GautssonMichael GreveTom O'ConnorLaura OldsJon HallidayPaul HobbsJeroen De DauwLupuleasa IonuțCooper LawtonTim NeilsonEric ScammellIgor KellerBen Glantonanul kumar sinhaDuncan OrrWill GlynnTyler HerrmannTomas SayderIan MunroJérôme BeaulieuNathan FishTaras BobrovytskyJeremy Vaskó RichárdBenjamin WatkinSebastian BirjoveanuAndrew HarcourtLuc RitchieNicholas GuyettJames Hinchcliffe12tone Chris BeachamZachary GidwitzNikita KiriyParker Andrew SchreiberSteve TrambertMario LoisAbigail Novickheino hulsey-vincentFionn Dmitri AfanasjevMarcel WardRichárd NagyfiAndrew WeirKabsMiłosz WierzbickiTendayi MawusheJannik OlbrichJake FishWr4thon Martin OttosenRobert HildebrandtAndy KobrePoker ChenKees Darko SperacPaul MoffatRobert ValdimarssonMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrik Daniel MunterAlex KnauthKasper Rob DawsonIan ReyesJames FowkesTom SayerLen Alan BandurkaBen HSimon PilkingtonDaniel KokotajloDiagon Andreas BlomqvistBertalan BodorDavid MorganZannheim Daniel Eickhardtlyon549 HD Ihor Mukha14zRobot Ivan Jason CherryIgor (Kerogi) Kostenkoib_ Thomas DingemanseStuart AlldrittAlexander BrownDevon BernardTed StokesJames HelmsJesper AnderssonJim TDeepFriedJif Chris DinantRaphaël LévyJohannes WalterMatt StantonGarrett MaringAnthony ChiuGhaith TarawnehJulian SchulzStellated HexahedronCaleb Scott ViteriClay UptonConor ComiconorMichael RoeschterGeorg GrassIsak Matthias HölzlJim RenneyEdison FranklinPiers CalderwoodKrzysztof DereckiMikhail TikhomirovRichard OttoMatt BrauerJaeson BookerMateusz KrzaczekArtem HoncharovMichael WaltersTomasz GlinieckiMihaly BaraszMark WoodwardRanzearNeil PalmereRajeen NabidChristian EppleClark SchaeferOlivier CoutuIestyn bleasdale-shepherdMojoExMachinaMarek BelskiEric EldardEric RogstadEric CarlsonCaleb LarsonBraden TisdaleMax ChiswickPhillip Brandelpatreon.com/robertskmiles
Sharing the Benefits of AI: The Windfall Clause Robert Miles AI Safety 2020-07-06 | AI might create enormous amounts of wealth, but how is it going to be distributed?The Paper: fhi.ox.ac.uk/wp-content/uploads/Windfall-Clause-Report.pdfThe Post: fhi.ox.ac.uk/windfallclauseWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesGladamasScott WorleyJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidPeter RolfChad JonesTeague LasserAndrew BlackledgeFrank MarsmanBrad BrookshireCam MacFarlaneJason HiseErik de BruijnAlec JohnsonClemens ArbesserLudwig SchubertBryce DaifukuAllen FaureEric JamesMatheson BayleyQeith Wreidjugettje dutchkingOwen Campbell-MooreAtzin Espino-MurnanePhil MoyerJacob Van BurenJonatan RIngvi GautssonMichael GreveJulius BrashTom O'ConnorShevis JohnsonLaura OldsJon HallidayPaul HobbsJeroen De DauwLupuleasa IonuțTim NeilsonEric ScammellIgor KellerBen Glantonanul kumar sinhaSean GibatDuncan OrrCooper LawtonWill GlynnTyler HerrmannTomas SayderIan MunroJérôme BeaulieuNathan FishTaras BobrovytskyJeremy Vaskó RichárdBenjamin WatkinEuclidean PlaneAndrew HarcourtLuc RitchieNicholas GuyettJames HinchcliffeOliver HabrykaChris BeachamZachary GidwitzNikita KiriyAndrew SchreiberDmitri AfanasjevMarcel WardAndrew WeirBen ArcherKabs Miłosz WierzbickiTendayi MawusheJannik OlbrichJake FishJussi MännistöWr4thon Martin OttosenArchy de BerkerAndy KobrePoker ChenKees Paul MoffatRobert ValdimarssonAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrik Daniel MunterAlex KnauthLeo Rob DawsonBryan EganRobert HildebrandtJames FowkesLen Alan BandurkaBen HTatiana PonomarevaMichael BatesSimon PilkingtonDaniel KokotajloFionn Diagon Andreas BlomqvistBertalan BodorDavid MorganBen SchultzZannheim Daniel Eickhardtlyon549 HD Ihor Mukha14zRobot Ivan Jason CherryIgor (Kerogi) Kostenkoib_ Thomas DingemanseStuart AlldrittAlexander BrownDevon BernardTed StokesJesper AnderssonJim TKasper DeepFriedJif Chris DinantRaphaël LévyMarko TopolnikJohannes WalterMatt StantonGarrett MaringMo HossnyAnthony ChiuFrank KurkaGhaith TarawnehJosh TrevisiolJulian SchulzStellated HexahedronCaleb Scott Viteri12tone Clay UptonBrent ODellConor ComiconorMichael RoeschterGeorg GrassIsak Matthias HölzlJim RenneyMichael V brownMartin HenriksenEdison FranklinDaniel SteelePiers CalderwoodKrzysztof DereckiMikhail TikhomirovRichárd NagyfiRichard OttoAlston SleetMatt BrauerJaeson BookerMateusz KrzaczekArtem HoncharovEvan WardMichael WaltersTomasz GlinieckiMihaly BaraszMark WoodwardRanzearNeil PalmereRajeen Nabidpatreon.com/robertskmiles
10 Reasons to Ignore AI Safety Robert Miles AI Safety 2020-06-04 | Why do some ignore AI Safety? Let's look at 10 reasons people give (adapted from Stuart Russell's list).Related Videos from Me:Why Would AI Want to do Bad Things? Instrumental Convergence: youtu.be/ZeecOKBus3QIntelligence and Stupidity: The Orthogonality Thesis: youtu.be/hEUO6pjwFOoPredicting AI: RIP Prof. Hubert Dreyfus: youtu.be/B6Oigy1i3W4A Response to Steven Pinker on AI: youtu.be/yQE9KAbFhNYRelated Videos from Computerphile:AI Safety: youtu.be/IB1OvoCNnWYGeneral AI Won't Want You To Fix its Code:youtu.be/4l7Is6vOAOAAI 'Stop Button' Problem: youtu.be/3TYT1QfdfsMProvably Beneficial AI - Stuart Russell: youtu.be/pARXQnX6QS8With thanks to my excellent Patreon supporters:patreon.com/robertskmilesGladamas James Scott WorleyJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidPeter RolfChad JonesFrank KurkaTeague LasserAndrew BlackledgeVignesh RavichandranJason HiseErik de BruijnClemens ArbesserLudwig SchubertBryce DaifukuAllen FaureEric JamesQeith Wreidjugettje dutchkingOwen Campbell-MooreAtzin Espino-MurnaneJacob Van BurenJonatan RIngvi GautssonMichael GreveJulius BrashTom O'ConnorShevis JohnsonLaura OldsJon HallidayPaul HobbsJeroen De DauwLupuleasa IonuțTim NeilsonEric ScammellIgor KellerBen Glantonanul kumar sinhaSean GibatDuncan OrrCooper LawtonWill GlynnTyler HerrmannTomas SayderIan MunroJérôme BeaulieuNathan FishTaras BobrovytskyJeremy Vaskó RichárdBenjamin WatkinSebastian BirjoveanuEuclidean PlaneAndrew HarcourtLuc RitchieNicholas GuyettJames HinchcliffeOliver HabrykaChris BeachamNikita Kiriyrobertvanduursen Dmitri AfanasjevMarcel WardAndrew WeirBen ArcherKabs Miłosz WierzbickiTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöWr4thon Martin OttosenArchy de BerkerAndy KobreBrian GillespiePoker ChenKees Darko SperacPaul MoffatAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrik Daniel MunterLeo Rob DawsonBryan EganRobert HildebrandtJames FowkesLen Alan BandurkaBen HTatiana PonomarevaMichael BatesSimon PilkingtonDaniel KokotajloFionn Diagon Parker LundRussell schoenAndreas BlomqvistBertalan BodorDavid MorganBen SchultzZannheim Daniel Eickhardtlyon549 HD Ihor Mukha14zRobot Ivan Jason CherryIgor (Kerogi) Kostenkoib_ Thomas DingemanseAlexander BrownDevon BernardTed StokesJesper AnderssonJim TKasper DeepFriedJif Daniel BartovicChris DinantRaphaël LévyMarko TopolnikJohannes WalterMatt StantonGarrett MaringMo HossnyAnthony ChiuGhaith TarawnehJosh TrevisiolJulian SchulzStellated HexahedronCaleb Scott Viteri12tone Nathaniel RaddinClay UptonBrent ODellConor ComiconorMichael RoeschterGeorg GrassIsak Matthias HölzlJim RenneyMichael V brownMartin HenriksenEdison FranklinDaniel SteelePiers CalderwoodKrzysztof DereckiZachary GidwitzMikhail Tikhomirovpatreon.com/robertskmiles
9 Examples of Specification Gaming Robert Miles AI Safety 2020-04-29 | AI systems do what you say, and it's hard to say exactly what you mean.Let's look at a list of real life examples of specification gaming!Related Videos from me:Reward Hacking: youtu.be/92qDfT8pENsReward Hacking Reloaded: youtu.be/46nsTFfsBucWhat Can We Do About Reward Hacking?: youtu.be/13tZ9Yia71cThe list: http://tinyurl.com/specification-gamingThe blogpost this video is based on: vkrakovna.wordpress.com/2018/04/02/specification-gaming-examples-in-aiThe newer blogpost that happened while I was making this video: deepmind.com/blog/article/Specification-gaming-the-flip-side-of-AI-ingenuity(Explosion graphic from videezy.com)Thanks to my wonderful patrons:patreon.com/robertskmilesGladamas James Steef Scott WorleyChad JonesChris CanalDavid ReidFrancisco TolmaskyFrank KurkaJake EhrlichJJ HepboinKellen laskMichael AndreggPedro A OrtegaPeter RolfSaid PolatTeague LasserAllen FaureBryce DaifukuClemens ArbesserEric JamesErik de BruijnJason Hisejugettje dutchkingLudwig SchubertQeith WreidAndrew Harcourtanul kumar sinhaBen GlantonBenjamin WatkinCooper LawtonDuncan OrrEric ScammellEuclidean PlaneIan MunroIgor KellerIngvi GautssonJames HinchcliffeJeroen De DauwJon HallidayJonatan RJulius BrashJérôme BeaulieuLaura OldsLuc RitchieLupuleasa IonuțMichael GreveNathan FishNicholas GuyettPaul HobbsSean GibatSebastian BirjoveanuShevis JohnsonTaras BobrovytskyTim NeilsonTom O'ConnorTomas SayderTyler HerrmannVaskó RichárdWill Glynn12tone 14zRobot Alan BandurkaAlexander BrownAnders ÖhrtAndreas BlomqvistAndrew WeirAndy KobreAnne KohlbrennerAnthony ChiuArchy de BerkerBen ArcherBen HBen SchultzBertalan BodorBrian GillespieBryan EganCaleb Chris DinantDaniel BartovicDaniel EickhardtDaniel KokotajloDaniel MunterDarko SperacDavid MorganDeepFriedJif Devon BernardDiagon Dmitri AfanasjevFionn Fraser CainGarrett MaringGhaith TarawnehHD Hendrik ib_ Igor (Kerogi) KostenkoIhor MukhaIvan James FowkesJannik OlbrichJason CherryJeremy Jesper AnderssonJim TJohannes WalterJosh TrevisiolJulian SchulzJussi MännistöKabs Kasper Kasper SchnackKees Klemen SlavicLeo lyon549 Marc PaulyMarcel WardMarco TiraboschiMarko TopolnikMartin OttosenMatt StantonMelisa KostrzewskiMichael BatesMichael KuhinicaMiłosz WierzbickiMo HossnyNathaniel RaddinOct todo22Owen Campbell-MooreParker LundPatrick HendersonPaul MoffatPoker ChenRob DawsonRobert Hildebrandtrobertvanduursen Robin ScharfRussell schoenScott ViteriSimon PilkingtonStellated HexahedronTatiana PonomarevaTed StokesTendayi MawusheThomas Dingemanse
Training AI Without Writing A Reward Function, with Reward Modelling Robert Miles AI Safety 2019-12-13 | How do you get a reinforcement learning agent to do what you want, when you can't actually write a reward function that specifies what that is?The paper: arxiv.org/pdf/1706.03741.pdfThe blogpost: openai.com/blog/deep-reinforcement-learning-from-human-preferencesThanks to my wonderful patrons:patreon.com/robertskmilesJamesGladamasSteefScott WorleyJordan MedinaSimon StrandgaardJJ HepboinPedro A OrtegaSaid PolatChris CanalJake EhrlichKellen laskFrancisco TolmaskyMichael AndreggDavid ReidRobert Daniel PickardPeter RolfChad JonesRichárd NagyfiJason HisePhil MoyerShevis JohnsonErik de BruijnAlec JohnsonClemens ArbesserLudwig SchubertBryce DaifukuAllen FaureEric JamesQeith WreidJonatan RIngvi GautssonMichael GreveJulius BrashTom O'ConnorRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwLupuleasa IonuțTim NeilsonEric ScammellIgor KellerBen Glantonanul kumar sinhaSean GibatCooper LawtonWill GlynnTyler HerrmannTomas SayderIan MunroJérôme BeaulieuNathan FishTaras BobrovytskyAnne BuitVaskó RichárdSebastian BirjoveanuEuclidean PlaneAndrew HarcourtDGJonorobertvanduursenDmitri AfanasjevMarcel WardAndrew WeirBen ArcherKabsMiłosz WierzbickiTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöWr4thonMartin OttosenArchy de BerkerMarc PaulyAndy KobreBrian GillespiePoker ChenKeesDarko SperacTrulsPaul MoffatAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfSeth BrothwellKasper SchnackKlemen SlavicPatrick HendersonOct todo22Melisa KostrzewskiHendrikDaniel MunterGraham HenryDuncan OrrBryan EganRobert HildebrandtJames FowkesAlan BandurkaBen HTatiana PonomarevaMichael BatesSimon PilkingtonDion Gerald BridgerPetr SmitalDaniel KokotajloFionnYuchong LiDiagonParker LundPaul EmmerichRussell schoenAndreas BlomqvistBertalan BodorDavid MorganJeremyBen SchultzZannheimDaniel Eickhardtlyon549HDIhor Mukha14zRobotIvanArne StrasserJason CherryIgor (Kerogi) KostenkoIsaac BoatesThomas DingemanseDavy KerAlexander BrownDevon BernardTed StokesJames HelmsMatheson Bayleypatreon.com/robertskmiles
AI That Doesnt Try Too Hard - Maximizers and Satisficers Robert Miles AI Safety 2019-08-23 | Powerful AI systems can be dangerous in part because they pursue their goals as strongly as they can. Perhaps it would be safer to have systems that don't aim for perfection, and stop at 'good enough'. How could we build something like that?Generating Fake YouTube comments with GPT-2: youtu.be/M6EXmoP5jX8Computerphile Videos:Unicorn AI: youtu.be/89A4jGvaaKkMore GPT-2, the 'writer' of Unicorn AI: youtu.be/p-6F4rhRYLQAI Language Models & Transformers: youtu.be/rURRYI66E54GPT-2: Why Didn't They Release It?: youtu.be/AJxLtdur5fcThe Deadly Truth of General AI?: youtu.be/tcdVC4e6EV4With thanks to my excellent Patreon supporters:patreon.com/robertskmilesScott WorleyJordan MedinaSimon StrandgaardJJ HepboinLupuleasa IonuțPedro A OrtegaSaid PolatChris CanalNicholas Kees DupuisJake EhrlichMark HechimKellen laskFrancisco TolmaskyMichael AndreggAlexandru DobreDavid ReidRobert Daniel PickardPeter RolfChad JonesTruthdocJamesRichárd NagyfiJason HisePhil MoyerShevis JohnsonAlec JohnsonClemens ArbesserLudwig SchubertBryce DaifukuAllen FaureEric JamesJonatan RIngvi GautssonMichael GreveJulius BrashTom O'ConnorErik de BruijnRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwTim NeilsonEric ScammellIgor KellerBen GlantonRobert Sokolowskianul kumar sinhaJérôme FrossardSean GibatCooper LawtonTyler HerrmannTomas SayderIan MunroJérôme BeaulieuTaras BobrovytskyAnne BuitTom MurphyVaskó RichárdSebastian BirjoveanuGladamasSylvain ChevalierDGJonoDmitri AfanasjevBrian SandbergMarcel WardAndrew WeirBen ArcherScott McCarthyKabsMiłosz WierzbickiTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöMr FantasticWr4thonMartin OttosenArchy de BerkerMarc PaulyJoshua PrattAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKeesDarko SperacTrulsPaul MoffatAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesSeth BrothwellClark MitchellKasper SchnackMichael HunterKlemen SlavicPatrick HendersonLong NguyenMelisa KostrzewskiHendrikDaniel MunterGraham HenryVolotatDuncan OrrMarin AldimirovBryan EganJames FowkesFrame ProblemsAlan BandurkaBenjamin HullTatiana PonomarevaAleksi MaunuMichael BatesSimon PilkingtonDion Gerald BridgerSteven CopeMarcos Alfredo NúñezPetr SmitalDaniel KokotajloFionnYuchong LiNathan FishDiagonParker LundRussell schoenAndreas BlomqvistBertalan BodorDavid MorganBen SchultzZannheimDaniel Eickhardtlyon549HDpatreon.com/robertskmiles
Is AI Safety a Pascals Mugging? Robert Miles AI Safety 2019-05-16 | An event that's very unlikely is still worth thinking about, if the consequences are big enough. What's the limit though?Do we have to devote all of our resources to any outcome that might give infinite payoffs, even if it seems basically impossible? Does the case for AI Safety rely on this kind of Pascal's Wager argument? Watch this video to find out that the answer to these questions is 'No'.Correction: At 6:34 the embedded video says 3^^^3 has 3.6 trillion digits, but that's actually only the size of 3^^4. 3^^^3 is enormously larger.The Alignment Newsletter Podcast: http://alignment-newsletter.libsyn.comRSS feed to put into apps: http://alignment-newsletter.libsyn(dot)com/rssWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesJason HiseJordan MedinaScott WorleyJJ HepboinPedro A OrtegaSaid PolatChris CanalNicholas Kees DupuisJake EhrlichMark HechimKellen laskFrancisco TolmaskyMichael AndreggJamesRichárd NagyfiPhil MoyerShevis JohnsonAlec JohnsonLupuleasa IonuțClemens ArbesserBryce DaifukuAllen FaureSimon StrandgaardJonatan RMichael GreveJulius BrashTom O'ConnorErik de BruijnRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwTim NeilsonEric ScammellIgor KellerBen GlantonRobert Sokolowskianul kumar sinhaJérôme FrossardSean GibatA.RussellCooper LawtonTyler HerrmannTomas SayderIan MunroJérôme BeaulieuGladamasSylvain ChevalierDGJonorobertvanduursenDmitri AfanasjevBrian SandbergMarcel WardAndrew WeirBen ArcherScott McCarthyKabsTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöMr FantasticWr4thonArchy de BerkerMarc PaulyJoshua PrattAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKeesTrulsPaul MoffatAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesSeth BrothwellBrian GoodrichKasper SchnackMichael HunterKlemen SlavicPatrick HendersonLong NguyenMelisa KostrzewskiHendrikDaniel MunterGraham HenryVolotatDuncan OrrBryan EganJames FowkesFrame ProblemsAlan BandurkaBenjamin HullDave TapleyTatiana PonomarevaAleksi MaunuMichael BatesSimon PilkingtonDion Gerald BridgerSteven CopePetr SmitalDaniel KokotajloJoshua DavisFionnTyler LaBeanRogerYuchong LiNathan FishDiagonGiancarlo Pacepatreon.com/robertskmiles
A Response to Steven Pinker on AI Robert Miles AI Safety 2019-03-31 | Steven Pinker wrote an article on AI for Popular Science Magazine, which I have some issues with.The article: popsci.com/robot-uprising-enlightenment-nowRelated:"The Orthogonality Thesis, Intelligence, and Stupidity" (youtu.be/hEUO6pjwFOo)"AI? Just Sandbox it... - Computerphile" (youtu.be/i8r_yShOixM)"Experts' Predictions about the Future of AI" (youtu.be/HOJ1NVtlnyQ)"Why Would AI Want to do Bad Things? Instrumental Convergence" (youtu.be/ZeecOKBus3Q)With thanks to my excellent Patreon supporters:patreon.com/robertskmilesJason HiseJordan MedinaScott WorleyJJ HepboinPedro A OrtegaSaid PolatChris CanalNicholas Kees DupuisJamesRichárd NagyfiPhil MoyerShevis JohnsonAlec JohnsonLupuleasa IonuțClemens ArbesserBryce DaifukuAllen FaureSimon StrandgaardJonatan RMichael GreveThe Guru Of VisionJulius BrashTom O'ConnorErik de BruijnRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwTim NeilsonEric ScammellIgor KellerBen GlantonRobert Sokolowskianul kumar sinhaJérôme FrossardSean GibatVolotatandrew RussellCooper LawtonGladamasSylvain ChevalierDGJonorobertvanduursenDmitri AfanasjevBrian SandbergMarcel WardAndrew WeirBen ArcherScott McCarthyKabsTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöMr FantasticWr4thonArchy de BerkerMarc PaulyJoshua PrattAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKeesDarko SperacTrulsPaul MoffatAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesSeth BrothwellBrian GoodrichClark MitchellKasper SchnackMichael HunterKlemen SlavicPatrick HendersonLong NguyenOct todo22Melisa KostrzewskiHendrikDaniel MunterGraham HenryDuncan OrrAndrew WalkerBryan Eganpatreon.com/robertskmiles
How to Keep Improving When Youre Better Than Any Teacher - Iterated Distillation and Amplification Robert Miles AI Safety 2019-03-11 | [2nd upload] AI systems can be trained using demonstrations from experts, but how do you train them to out-perform those experts? Can this still be done even without clear win/loss criteria? And how do you do it safely?This video was based on work including:"Supervising strong learners by amplifying weak experts" by Paul Christiano, Buck Shlegeris, Dario Amodei (arxiv.org/abs/1810.08575)openai.com/blog/amplifying-ai-trainingalignmentforum.org/s/EmDuGeRw749sD3GKdai-alignment.com/iterated-distillation-and-amplification-157debfd1616With thanks to my wonderful Patrons: ( patreon.com/robertskmiles )Steef Jason StrackJordan MedinaJason HiseScott WorleyJJ HepboinPedro A OrtegaSaid PolatChris CanalNicholas Kees DupuisJames Richárd NagyfiPhil MoyerAlec JohnsonClemens ArbesserBryce DaifukuSimon StrandgaardJonatan RMichael GreveThe Guru Of VisionVolodymyr David TjäderJulius BrashTom O'ConnorErik de BruijnRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwTim NeilsonEric ScammellIgor KellerBen GlantonRobert Sokolowskianul kumar sinhaJérôme FrossardSean GibatSun Sunandrew RussellCooper LawtonGladamas Sylvain ChevalierDGJono robertvanduursen Dmitri AfanasjevBrian SandbergEinar UelandMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabs Kabs KabsTendayi MawusheJannik OlbrichAnne KohlbrennerBjorn NybladJussi MännistöMr FantasticWr4thon Archy de BerkerMarc PaulyJoshua PrattShevis JohnsonAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKees Darko SperacTruls Paul MoffatJelle LangenAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesShawn HartsockSeth BrothwellBrian GoodrichClark MitchellKasper SchnackMichael HunterKlemen SlavicPatrick HendersonLong NguyenOct todo22Melisa KostrzewskiHendrik Daniel MunterGraham HenryDuncan Orr
Why Not Just: Think of AGI Like a Corporation? Robert Miles AI Safety 2018-12-23 | Corporations are kind of like AIs, if you squint. How hard do you have to squint though, and is it worth it?In this video we ask: Are corporations artificial general superintelligences?Related:"What can AGI do? I/O and Speed" (youtu.be/gP4ZNUHdwp8)"Why Would AI Want to do Bad Things? Instrumental Convergence" (youtu.be/ZeecOKBus3Q)Media Sources:"SpaceX - How Not to Land an Orbital Rocket Booster" (youtu.be/bvim4rsNHkQ)Undertale - TurbosnailClerks (1994)Zootopia (2016)AlphaGo (2017)Ready Player One (2018)With thanks to my excellent Patreon supporters:patreon.com/robertskmilesJordan MedinaJason HisePablo EderScott WorleyJJ HepboinPedro A OrtegaJames McCuenRichárd NagyfiPhil MoyerAlec JohnsonBobby ColdClemens ArbesserSimon StrandgaardJonatan RMichael GreveThe Guru Of VisionDavid TjäderJulius BrashTom O'ConnorErik de BruijnRobin GreenLaura OldsJon HallidayPaul HobbsJeroen De DauwTim NeilsonEric ScammellIgor KellerBen GlantonRobert SokolowskiJérôme FrossardSean GibatSylvain ChevalierDGJonorobertvanduursenScott StevensDmitri AfanasjevBrian SandbergMarcel WardAndrew WeirBen ArcherScott McCarthyKabs Kabs KabsTendayi MawusheJannik OlbrichAnne KohlbrennerJussi MännistöMr FantasticWr4thonDave TapleyArchy de BerkerKevinMarc PaulyJoshua PrattGunnar GuðvarðarsonShevis JohnsonAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKeesDarko SperacTrulsPaul MoffatAnders ÖhrtLupuleasa IonuțMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesShawn HartsockSeth BrothwellBrian GoodrichMichael S McReynoldsClark MitchellKasper SchnackMichael HunterKlemen SlavicPatrick Hendersonpatreon.com/robertskmiles
Safe Exploration: Concrete Problems in AI Safety Part 6 Robert Miles AI Safety 2018-09-21 | To learn, you need to try new things, but that can be risky. How do we make AI systems that can explore safely?Playlist of the series so far: youtube.com/playlist?list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778The paper, 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdfAI Safety Gridworlds: youtu.be/CGTkoUidQ8IWhy Would AI Want to do Bad Things? Instrumental Convergence: youtu.be/ZeecOKBus3QScalable Supervision: Concrete Problems in AI Safety Part 5: youtu.be/nr1lHuFeq5wThe Evolved Radio and its Implications for Modelling the Evolution of Novel Sensors: https://people.duke.edu/~ng46/topics/evolved-radio.pdfWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesJason HiseSteefJason StrackStefan SkilesJordan MedinaScott WorleyJJ HepboinAlex FlintPedro A OrtegaJames McCuenRichárd NagyfiAlec JohnsonClemens ArbesserSimon StrandgaardJonatan RMichael GreveThe Guru Of VisionAlexander Hartvig NielsenVolodymyrDavid TjäderJulius BrashTom O'ConnorVille AhlgrenErik de BruijnRobin GreenMaksym TaranLaura OldsJon HallidayBobby ColdPaul HobbsJeroen De DauwTim NeilsonEric Scammellchristopher dasenbrockIgor KellerBen GlantonRobert SokolowskiVlad DJérôme FrossardLupuleasa IonuțSylvain ChevalierDGJonorobertvanduursenScott StevensDmitri AfanasjevBrian SandbergEinar UelandMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabsPhil MoyerTendayi MawusheAnne KohlbrennerBjorn NybladJussi MännistöMr FantasticMatanya LoewenthalWr4thonDave TapleyArchy de BerkerPablo EderKevinMarc PaulyJoshua PrattGunnar GuðvarðarsonShevis JohnsonAndy KobreManuel WeichselbaumBrian GillespieMartin WindPeggy YouellPoker ChenKeesDarko SperacPaul MoffatJelle LangenLars ScholzAnders ÖhrtMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesGladamasShawn HartsockSeth BrothwellBrian GoodrichMichael S McReynoldspatreon.com/robertskmilesMedia Sources:"DashCam Russia - Crazy Drivers and Car Crashes 2018" (youtu.be/h50TQ3i9k5I)Optimist Prime"Hapless Boston Dynamics robot in shelf-stacking fail" (youtu.be/JzlsvFN_5HI)"The Simpsons - Bart Gets Famous" (c) Fox 1994"Donald Duck - Cured Duck" (c) Disney 1945"Vase Breaking Slow Motion" (youtu.be/IJNPc_anP7U)"Fastest quadcopter i've ever flown + Most Destructive Crash" (youtu.be/OKT4cx7UKsk)"An athlete uses physics to shatter world records - Asaf Bar-Yosef" (youtu.be/RaGUW1d0w8g)"Uber self-driving car crash in Tempe, Arizona" (youtu.be/XtTB8hTgHbM)"Quadcopter Fx Simulator" (youtu.be/-6si8WkRtaY)"Fallout - New Vegas by progamingwithed in 24:00 - AGDQ 2017 - Part 59" (youtu.be/nuzDif16_nc)"Far Cry 5 out of 5 Physics Simulation" (youtu.be/4My0Bt30pX0)
Friend or Foe? AI Safety Gridworlds extra bit Robert Miles AI Safety 2018-06-24 | The last video about the AI Safety Gridworlds paper. How does an agent detect and adapt to friendly and adversarial intentions in the environment?The previous video: youtu.be/CGTkoUidQ8IThe Computerphile video: youtube.com/watch?v=eElfR_BnL5kThe EXTRA BITS video, with more detail: youtube.com/watch?v=py5VRagG6t8The paper: arxiv.org/pdf/1711.09883.pdfThe GitHub repos: github.com/deepmind/ai-safety-gridworlds With thanks to my excellent Patreon supporters:patreon.com/robertskmilesJason HiseSteef Cooper LawtonJason StrackChad JonesStefan SkilesJordan MedinaManuel WeichselbaumScott WorleyJJ HepboinAlex FlintJustin CourtrightPedro A OrtegaJames McCuenRichárd NagyfiVille AhlgrenAlec JohnsonClement ChirisSimon StrandgaardJoshua RichardsonJonatan RMichael GreveThe Guru Of VisionAlexander Hartvig NielsenVolodymyr David TjäderJulius BrashTom O'ConnorGunnar GuðvarðarsonShevis JohnsonErik de BruijnRobin GreenAlexei VasilkovMaksym TaranLaura OldsJon HallidayRobert WernerPaul HobbsJeroen De DauwEnrico RosTim NeilsonEric Scammellchristopher dasenbrockIgor KellerMorten JelleBen GlantonRobert SokolowskiVlad DWilliam HendleyDGJono robertvanduursen Scott StevensEmilio AlvarezDmitri AfanasjevBrian SandbergEinar UelandMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabs Phil Tendayi MawusheAnne KohlbrennerJake FishBjorn NybladJussi MännistöMr FantasticMatanya LoewenthalWr4thon Dave TapleyArchy de BerkerKevin Marc PaulyJoshua PrattAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKees Darko SperacPaul MoffatJelle LangenLars ScholzAnders ÖhrtLupuleasa IonuțMarco TiraboschiMichael KuhinicaFraser CainRobin ScharfOren MilmanJohn ReesShawn HartsockSeth Brothwellpatreon.com/robertskmiles
AI Safety Gridworlds Robert Miles AI Safety 2018-05-25 | Got an AI safety idea? Now you can test it out! A recent paper from DeepMind sets out some environments for evaluating the safety of AI systems, and the code is on GitHub.The Computerphile video: youtube.com/watch?v=eElfR_BnL5kThe EXTRA BITS video, with more detail: youtube.com/watch?v=py5VRagG6t8The paper: arxiv.org/pdf/1711.09883.pdfThe GitHub repos: github.com/deepmind/ai-safety-gridworlds patreon.com/robertskmilesWith thanks to my wonderful Patreon supporters:- Jason Hise- Steef - Cooper Lawton- Jason Strack- Chad Jones- Stefan Skiles- Jordan Medina- Manuel Weichselbaum- Scott Worley- JJ Hepboin- Alex Flint- Justin Courtright- James McCuen- Richárd Nagyfi- Ville Ahlgren- Alec Johnson- Simon Strandgaard- Joshua Richardson- Jonatan R- Michael Greve- The Guru Of Vision- Fabrizio Pisani- Alexander Hartvig Nielsen- Volodymyr - David Tjäder- Paul Mason- Ben Scanlon- Julius Brash- Mike Bird- Tom O'Connor- Gunnar Guðvarðarson- Shevis Johnson- Erik de Bruijn- Robin Green- Alexei Vasilkov- Maksym Taran- Laura Olds- Jon Halliday- Robert Werner- Paul Hobbs- Jeroen De Dauw- Enrico Ros- Tim Neilson- Eric Scammell- christopher dasenbrock- Igor Keller- William Hendley- DGJono - robertvanduursen - Scott Stevens- Michael Ore- Dmitri Afanasjev- Brian Sandberg- Einar Ueland- Marcel Ward- Andrew Weir- Taylor Smith- Ben Archer- Scott McCarthy- Kabs Kabs- Phil - Tendayi Mawushe- Gabriel Behm- Anne Kohlbrenner- Jake Fish- Bjorn Nyblad- Jussi Männistö- Mr Fantastic- Matanya Loewenthal- Wr4thon - Dave Tapley- Archy de Berker- Kevin - Marc Pauly- Joshua Pratt- Andy Kobre- Brian Gillespie- Martin Wind- Peggy Youell- Poker Chen- pmilian - Kees - Darko Sperac- Paul Moffat- Jelle Langen- Lars Scholz- Anders Öhrt- Lupuleasa Ionuț- Marco Tiraboschi- Peter Kjeld Andersen- Michael Kuhinica- Fraser Cain- Robin Scharf- Oren Milman
Experts Predictions about the Future of AI Robert Miles AI Safety 2018-03-31 | When will AI systems surpass human performance? I don't know, do you? No you don't. Let's see what 352 top AI researchers think.[CORRECTION: I mistakenly stated that the survey was before AlphaGo beat Lee Sedol. The 12 year prediction was for AI to outperform humans *after having only played as many games as a human plays in their lifetime*]The paper: arxiv.org/pdf/1705.08807.pdfThe blogpost which has lots of nice data visualisations: aiimpacts.org/2016-expert-survey-on-progress-in-aiThe Instrumental Convergence video: youtube.com/watch?v=ZeecOKBus3QThe Negative Side Effects video: youtube.com/watch?v=lqJUIqZNzP8With thanks to my excellent Patrons at patreon.com/robertskmiles :Jason HiseSteef Jason StrackChad JonesStefan SkilesJordan MedinaManuel Weichselbaum1RV34 Scott WorleyJJ HepboinAlex FlintJames McCuenRichárd NagyfiVille AhlgrenAlec JohnsonSimon StrandgaardJoshua RichardsonJonatan RMichael GreveThe Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTom O'ConnorGunnar GuðvarðarsonShevis JohnsonErik de BruijnRobin GreenAlexei VasilkovMaksym TaranLaura OldsJon HallidayRobert WernerPaul HobbsJeroen De DauwKonsta William HendleyDGJono robertvanduursen Scott StevensMichael OreDmitri AfanasjevBrian SandbergEinar UelandMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabs KabsPhil Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishBjorn NybladJussi MännistöMr FantasticMatanya LoewenthalWr4thon Dave TapleyArchy de BerkerKevin Vincent SandersMarc PaulyAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKees Darko SperacPaul MoffatNoel KocherilJelle LangenLars Scholz
Why Would AI Want to do Bad Things? Instrumental Convergence Robert Miles AI Safety 2018-03-24 | How can we predict that AGI with unknown goals would behave badly by default?The Orthogonality Thesis video: youtube.com/watch?v=hEUO6pjwFOoInstrumental Convergence: arbital.com/p/instrumental_convergenceOmohundro 2008, Basic AI Drives: selfawaresystems.files.wordpress.com/2008/01/ai_drives_final.pdfWith thanks to my excellent Patrons at patreon.com/robertskmiles :Jason HiseSteef Jason StrackChad JonesStefan SkilesJordan MedinaManuel Weichselbaum1RV34 Scott WorleyJJ HepboinAlex FlintJames McCuenRichárd NagyfiVille AhlgrenAlec JohnsonSimon StrandgaardJoshua RichardsonJonatan RMichael GreveThe Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTom O'ConnorGunnar GuðvarðarsonShevis JohnsonErik de BruijnRobin GreenAlexei VasilkovMaksym TaranLaura OldsJon HallidayRobert WernerPaul HobbsJeroen De DauwKonsta William HendleyDGJono robertvanduursen Scott StevensMichael OreDmitri AfanasjevBrian SandbergEinar UelandMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabs KabsPhil Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishBjorn NybladJussi MännistöMr FantasticMatanya LoewenthalWr4thon Dave TapleyArchy de BerkerKevin Vincent SandersMarc PaulyAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKees Darko SperacPaul MoffatNoel KocherilJelle LangenLars Scholz
Superintelligence Mod for Civilization V Robert Miles AI Safety 2018-02-13 | Let's play this new mod for Civ 5 that makes AGI an available technology!Can we guide humanity to a utopian AI future, or will we destroy ourselves?Download the mod here: steamcommunity.com/sharedfiles/filedetails/?id=1215263272or here if you have the Brave New World DLC: http://steamcommunity.com/sharedfiles/filedetails/?id=1217587000With special thanks to Dr Shahar Avin: cser.ac.uk/team/shahar-avinThe long version: youtu.be/SBwOJMZtJaorockpapershotgun.com/2018/02/06/the-creator-of-the-civilization-v-superintelligence-mod-on-ai-safetytheverge.com/2018/1/5/16853628/civilization-v-ai-risk-mod-cserWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesJason HiseSteef Jason StrackChad JonesStefan SkilesJordan MedinaManuel Weichselbaum1RV34 Scott WorleyJJ HepboinJames McCuenRichárd NagyfiVille AhlgrenAlec JohnsonTrevor Alexander NestorClement ChirisSimon StrandgaardJoshua RichardsonJonatan RMichael GreveThe Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTom O'ConnorGunnar GuðvarðarsonShevis JohnsonErik de BruijnRobin GreenAlexei VasilkovMaksym TaranLaura OldsJon HallidayRobert WernerPaul HobbsJeroen De DauwRoman NekhoroshevKonsta William HendleyDGJono robertvanduursen Scott StevensEmilio AlvarezMichael OreDmitri AfanasjevBrian SandbergEinar UelandLo RezMarcel WardAndrew WeirTaylor SmithBen ArcherScott McCarthyKabs KabsPhil Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishBjorn NybladStefan LaurieJussi MännistöCameron KinselMatanya LoewenthalWr4thon Dave TapleyArchy de BerkerKevin Vincent SandersMarc PaulyAndy KobreBrian GillespieMartin WindPeggy YouellPoker ChenKees BoonDarko SperacPaul MoffatNoel KocherilJelle Langenpatreon.com/robertskmiles
Intelligence and Stupidity: The Orthogonality Thesis Robert Miles AI Safety 2018-01-11 | Can highly intelligent agents have stupid goals?A look at The Orthogonality Thesis and the nature of stupidity.The 'Stamp Collector' Computerphile video: youtube.com/watch?v=tcdVC4e6EV4My other Computerphile videos: youtube.com/watch?v=GSIDS_lvRv4&list=PLqL14ZxTTA4fRMts7Af2G8t4Rp17e8MdSKatie Byrne's Channel: youtube.com/channel/UCwA5wd50HYLa-ZY80LSAg3wChad Jones' Channel: youtube.com/user/cjone150 patreon.com/robertskmilesWith thanks to my wonderful Patreon supporters:- Steef - Sara Tjäder- Jason Strack- Chad Jones- Stefan Skiles- Ziyang Liu- Jordan Medina- Jason Hise- Manuel Weichselbaum- 1RV34 - James McCuen- Richárd Nagyfi- Ammar Mousali- Scott Zockoll- Ville Ahlgren- Alec Johnson- Simon Strandgaard- Joshua Richardson- Jonatan R- Michael Greve- robertvanduursen - The Guru Of Vision- Fabrizio Pisani- Alexander Hartvig Nielsen- Volodymyr - David Tjäder- Paul Mason- Ben Scanlon- Julius Brash- Mike Bird- Tom O'Connor- Gunnar Guðvarðarson- Shevis Johnson- Erik de Bruijn- Robin Green- Alexei Vasilkov- Maksym Taran- Laura Olds- Jon Halliday- Robert Werner- Roman Nekhoroshev- Konsta - William Hendley- DGJono - Matthias Meger- Scott Stevens- Emilio Alvarez- Michael Ore- Dmitri Afanasjev- Brian Sandberg- Einar Ueland- Lo Rez- Marcel Ward- Andrew Weir- Taylor Smith- Ben Archer- Scott McCarthy- Kabs Kabs- Phil - Tendayi Mawushe- Gabriel Behm- Anne Kohlbrenner- Jake Fish- Bjorn Nyblad- Stefan Laurie- Jussi Männistö- Cameron Kinsel- Matanya Loewenthal- Wr4thon - Dave Tapley- Archy de Berker- Kevin - Vincent Sanders- Marc Pauly- Andy Kobre- Brian Gillespie- Martin Wind- Peggy Youell- Poker Chenpatreon.com/robertskmiles
Scalable Supervision: Concrete Problems in AI Safety Part 5 Robert Miles AI Safety 2017-11-29 | Why can't we just have humans overseeing our AI systems?The Concrete Problems in AI Safety Playlist: youtube.com/playlist?list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778 Previous Video: youtube.com/watch?v=13tZ9Yia71cThe Computerphile video: youtube.com/watch?v=9nktr1MgS-A The paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdf patreon.com/robertskmilesWith thanks to my wonderful Patreon supporters:- Steef - Sara Tjäder- Jason Strack- Chad Jones- Stefan Skiles- Ziyang Liu- Jordan Medina- Jason Hise- Heavy Empty- Manuel Weichselbaum- James McCuen- Richárd Nagyfi- Ammar Mousali- Scott Zockoll- Charles Miller- Joshua Richardson- Jonatan R- Michael Greve- robertvanduursen - The Guru Of Vision- Fabrizio Pisani- Alexander Hartvig Nielsen- Volodymyr - David Tjäder- Paul Mason- Ben Scanlon- Julius Brash- Mike Bird- Taylor Winning- Ville Ahlgren- Johannes David- Andrew Pearce- Gunnar Guðvarðarson- Shevis Johnson- Erik de Bruijn- Robin Green- Alexei Vasilkov- Roman Nekhoroshev- Peggy Youell- Konsta - William Hendley- Almighty Dodd- DGJono - Matthias Meger- Scott Stevens- Emilio Alvarez- Michael Ore- Robert Bridges- Dmitri Afanasjev- Brian Sandberg- Einar Ueland- Lo Rez- Stephen Paul- Marcel Ward- Andrew Weir- Pontus Carlsson- Taylor Smith- Ben Archer- Ivan Pochesnev- Scott McCarthy- Kabs Kabs Kabs- Phil - Christopher Askin- Tendayi Mawushe- Gabriel Behm- Anne Kohlbrenner- Jake Fish- David Rasmussen- Bjorn Nyblad- Stefan Laurie- Tom O'Connor- pmilian - Jussi Männistö- Cameron Kinsel- Matanya Loewenthal- Wr4thon - Dave Tapley- Archy de Berker- Kevin - Vincent Sanders- Marc Pauly- Andy Kobre- Brian Gillespiepatreon.com/robertskmiles
AI Safety at EAGlobal2017 Conference Robert Miles AI Safety 2017-11-16 | I attended a charity conference to learn about AI Safety!Correction: Alan Dafoe is funded by a grant from the Open Philanthropy Project, but does not work for them.The conference's YouTube channel: youtube.com/channel/UCEfASxwPxzsHlG5Rf1-4K9wThe Website: eaglobal.org/events/ea-global-2017-ukJobs at FHI: fhi.ox.ac.uk/vacanciesMy Concrete Problems in AI Safety series: youtube.com/watch?v=lqJUIqZNzP8&list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778With thanks to my Patrons! (patreon.com/robertskmiles)Steef Sara TjäderJason StrackChad JonesStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseHeavy EmptyJames McCuenRichárd NagyfiAmmar MousaliScott ZockollCharles MillerJoshua RichardsonJonatan RMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningVille AhlgrenJohannes DavidAndrew PearceGunnar GuðvarðarsonShevis JohnsonErik de BruijnRobin GreenRoman NekhoroshevPeggy YouellKonsta William HendleyAdam DoddDGJono Matthias MegerScott StevensMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezStephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs Kabs KabsPhil Christopher AskinTendayi MawusheGabriel BehmAnne KohlbrennerJake FishDavid RasmussenFilip Bjorn NybladStefan LaurieTom O'Connorpmilian Jussi MännistöCameron KinselMatanya LoewenthalWr4thon Dave TapleyArchy de Berkerpatreon.com/robertskmiles
AI learns to Create ̵K̵Z̵F̵ ̵V̵i̵d̵e̵o̵s̵ Cat Pictures: Papers in Two Minutes #1 Robert Miles AI Safety 2017-10-29 | Some beautiful new GAN results have been published, so let's have a quick look at the pretty pictures.More AI Safety coming soon of course.With apologies to Two Minute Papers: youtube.com/user/keeroyzThe Computerphile video: youtube.com/watch?v=Sw9r8CL98N0The Paper "Unsupervised Representation Learning With Deep Convolutional GANs": arxiv.org/pdf/1511.06434.pdfThe Paper "Progressive Growing Of Gans For Improved Quality, Stability, And Variation": http://research.nvidia.com/sites/default/files/pubs/2017-10_Progressive-Growing-of//karras2017gan-paper.pdfThe AMAZING video for that paper: youtube.com/watch?v=XOxxPcy5Gr4 With thanks to my excellent Patreon supporters:patreon.com/robertskmilesSteef Sara TjäderJason StrackChad JonesStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseDavid RasmussenHeavy EmptyJames McCuenRichárd NagyfiAmmar MousaliScott ZockollCharles MillerJoshua RichardsonJonatan RMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningVille AhlgrenRoman NekhoroshevPeggy YouellKonsta William HendleyAlmighty DoddDGJono Matthias MegerScott StevensMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezStephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs Kabs Kabs KabsPhil Christopher AskinTendayi MawusheGabriel BehmAnne KohlbrennerJake FishFilip Bjorn NybladStefan LaurieTom O'Connorpmilian Jussi MännistöCameron KinselMatanya LoewenthalWr4thon Dave TapleyArchy de Berkerpatreon.com/robertskmiles
What can AGI do? I/O and Speed Robert Miles AI Safety 2017-10-17 | Suppose we make an algorithm that implements general intelligence as well as the brain. What could that system do?It might have better input and output than a human, and probably could be run faster...The Computerphile video: youtube.com/watch?v=tcdVC4e6EV4The paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdfThey're Made Out Of Meat: youtube.com/watch?v=7tScAyNaRdQThe Slow Mo Guys' Channel: youtube.com/user/theslowmoguysWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesSteef Sara TjäderJason StrackChad JonesStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseDavid RasmussenHeavy EmptyJames McCuenRichárd NagyfiAmmar MousaliScott ZockollCharles MillerJoshua RichardsonJonatan RØystein FlygtMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningVille AhlgrenRoman NekhoroshevPeggy YouellKonstantin ShabashovWilliam HendleyAdam DoddDGJono Matthias MegerScott StevensMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezStephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs Phil Christopher Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishJennifer Autumn LathamFilip Bjorn NybladStefan LaurieTom O'Connorpmilian Jussi MännistöCameron KinselMatanya LoewenthalWr4thon Dave TapleyArchy de Berkerpatreon.com/robertskmiles
What Can We Do About Reward Hacking?: Concrete Problems in AI Safety Part 4 Robert Miles AI Safety 2017-09-24 | Three different approaches that might help to prevent reward hacking.New Side Channel with no content yet!: youtube.com/channel/UC4qH2AHly_RSRze1bUqSSNwWhere do we go now?: youtube.com/watch?v=vYhErnZdnsoPrevious Video in the series: youtu.be/46nsTFfsBucThe Concrete Problems in AI Safety Playlist: youtube.com/playlist?list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778 The Computerphile video: youtube.com/watch?v=4l7Is6vOAOAThe paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdf With thanks to my excellent Patreon supporters:patreon.com/robertskmilesSteef Sara TjäderJason StrackChad JonesStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseDavid RasmussenHeavy EmptyJames McCuenRichárd NagyfiAmmar MousaliScott ZockollCharles MillerJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniA Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningRoman NekhoroshevPeggy YouellKonstantin ShabashovDodd AlmightyDGJono Matthias MegerScott StevensEmilio AlvarezMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehne Stephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs Kabs KabsPhil Philip AlexanderChristopher Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishJennifer Autumn LathamFilip Bjorn NybladStefan LaurieTom O'ConnorKrethys PiotrekM Jussi MännistöMatanya LoewenthalWr4thon
Reward Hacking Reloaded: Concrete Problems in AI Safety Part 3.5 Robert Miles AI Safety 2017-08-29 | Goodhart's Law, Partially Observed Goals, and Wireheading: some more reasons for AI systems to find ways to 'cheat' and get more reward than we intended.The Concrete Problems in AI Safety Playlist: youtube.com/playlist?list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778 Previous Video: youtube.com/watch?v=92qDfT8pENsThe Computerphile video: youtube.com/watch?v=9nktr1MgS-A The paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdf SethBling's channel: youtube.com/user/sethblingWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesSteef Sara TjäderJason StrackChad JonesIchiro DohiStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseDavid RasmussenJames McCuenRichárd NagyfiAmmar MousaliScott ZockollCharles MillerJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningRoman NekhoroshevPeggy YouellKonstantin ShabashovAlmighty DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehne Stephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs KabsPhil Philip AlexanderChristopher Tendayi MawusheGabriel BehmAnne Kohlbrenner
The other Killer Robot Arms Race Elon Musk should worry about Robert Miles AI Safety 2017-08-22 | Elon Musk is in the news, talking to the UN about autonomous weapons. This seems like a good time to explain one area where we don't quite agree about AI Safety.The Article: http://www.independent.co.uk/news/science/killer-robots-arms-race-tesla-elon-musk-and-google-mustafa-suleyman-un-autonomous-weapons-a7903906.htmlThe clip at 2:54 is from a Y Combinator interview: "Elon Musk : How to Build the Future": youtu.be/tnBQmEqBCY0With thanks to my excellent Patreon supporters:patreon.com/robertskmilesSteef Sara TjäderJason StrackChad JonesIchiro DohiStefan SkilesKatie ByrneZiyang LiuJordan MedinaKyle ScottJason HiseDavid RasmussenJames McCuenRichárd NagyfiAmmar MousaliScott ZockollJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdTaylor WinningPeggy YouellKonstantin ShabashovAlmighty DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehne Stephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabs Kabs Phil Philip AlexanderChristopher Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishJennifer Autumn LathamFilip Bjorn NybladStefan LaurieTom O'ConnorKrethys
Reward Hacking: Concrete Problems in AI Safety Part 3 Robert Miles AI Safety 2017-08-12 | Sometimes AI can find ways to 'cheat' and get more reward than we intended by doing something unexpected. The Concrete Problems in AI Safety Playlist: youtube.com/playlist?list=PLqL14ZxTTA4fEp5ltiNinNHdkPuLK4778 The Computerphile video: youtube.com/watch?v=9nktr1MgS-A The paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdf SethBling's channel: youtube.com/user/sethblingWith thanks to my excellent Patreon supporters:patreon.com/robertskmilesJordan MedinaFHI's own Kyle ScottJason HiseDavid RasmussenJames McCuenRichárd NagyfiAmmar MousaliJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr David TjäderPaul MasonBen ScanlonJulius BrashMike BirdPeggy YouellKonstantin ShabashovAlmighty DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehne Stephen PaulMarcel WardAndrew WeirPontus CarlssonTaylor SmithBen ArcherIvan PochesnevScott McCarthyKabilan Kabilan Kabilan KabilanPhil Philip AlexanderChristopher Tendayi MawusheGabriel BehmAnne KohlbrennerJake FishJennifer Autumn Latham
Why Not Just: Raise AI Like Kids? Robert Miles AI Safety 2017-07-22 | Newly made Artificial General Intelligences are basically like children, right? So we already know we can teach them how to behave, right? Wrong.References to this Computerphile video: youtube.com/watch?v=tcdVC4e6EV4and this paper: intelligence.org/files/ValueLearningProblem.pdfThanks to my amazing Patreon Supporters:Sara TjäderJason StrackChad JonesIchiro DohiStefan SkilesKatie ByrneZiyang LiuJordan MedinaJames McCuenJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr Peggy YouellKonstantin ShabashovAlmighty DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehnepatreon.com/robertskmiles
Empowerment: Concrete Problems in AI Safety part 2 Robert Miles AI Safety 2017-07-09 | Maybe AI systems would be safer if they avoid gaining too much control over their environment? How might that work?This is a follow-up to this earlier video: youtu.be/lqJUIqZNzP8The paper 'Concrete Problems in AI Safety': arxiv.org/pdf/1606.06565.pdfA book chapter about Empowerment: arxiv.org/pdf/1310.1863.pdfProf Brailsford's Information Theory Videos: youtube.com/watch?v=Lto-ajuqW3w&list=PLzH6n4zXuckpKAj1_88VS-8Z6yn9zX_P6Thanks to my amazing Patreon Supporters:Sara TjäderJason StrackChad JonesIchiro DohiStefan SkilesKatie ByrneZiyang LiuJordan MedinaJames McCuenJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr Peggy YouellKonstantin ShabashovAlmighty DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo RezC3POehnepatreon.com/robertskmiles
Avoiding Positive Side Effects: Concrete Problems in AI Safety part 1.5 Robert Miles AI Safety 2017-06-25 | This is a follow-up to this earlier video: youtu.be/lqJUIqZNzP8There's another problem with minimising side effects...The paper we're looking at: arxiv.org/pdf/1606.06565.pdfAI Safety Careers advice: 80000hours.org/problem-profiles/positively-shaping-artificial-intelligenceThanks to my amazing Patreon Supporters:Chad JonesIchiro DohiStefan SkilesKatie ByrneZiyang LiuJames McCuenJoshua RichardsonFabian ConsiglioJonatan RØystein FlygtBjörn MostenMichael Greverobertvanduursen The Guru Of VisionFabrizio PisaniAlexander Hartvig NielsenVolodymyr Peggy YouellKonstantin ShabashovThe DoddDGJono Matthias MegerScott StevensEmilio AlvarezBenjamin Aaron DegenhartMichael OreRobert BridgesDmitri AfanasjevBrian SandbergEinar UelandLo Rezpatreon.com/robertskmiles
Avoiding Negative Side Effects: Concrete Problems in AI Safety part 1 Robert Miles AI Safety 2017-06-18 | We can expect AI systems to accidentally create serious negative side effects - how can we avoid that?The first of several videos about the paper "Concrete Problems in AI Safety".Earlier on Computerphile: youtube.com/watch?v=AjyM-f8rDpgThe Stop Button Problem: youtu.be/3TYT1QfdfsMRead the full paper here: arxiv.org/abs/1606.06565Thanks again to all my wonderful Patreon Supporters:- Chad Jones- Ichiro Dohi- Stefan Skiles- Katie Byrne- Ziyang Liu- Joshua Richardson- Fabian Consiglio- Jonatan R- Øystein Flygt- Björn Mosten- Michael Greve- robertvanduursen - The Guru Of Vision- Fabrizio Pisani- Alexander Hartvig Nielsen- Volodymyr - Peggy Youell- Konstantin Shabashov- Adam Dodd- DGJono - Matthias Meger- Scott Stevens- Emilio Alvarez- Benjamin Aaron Degenhart- Michael Ore- Robert Bridges- Dmitri Afanasjevpatreon.com/robertskmiles
Are AI Risks like Nuclear Risks? Robert Miles AI Safety 2017-06-10 | Concerns about AI cover a really wide range of possible problems. Can we make progress on several of these problems at once?With thanks to my Patreon supporters:- Ichiro Dohi- Stefan Skiles- Chad Jones- Joshua Richardson- Fabian Consiglio- Jonatan R- Øystein Flygt- Björn Mosten- Michael Greve- robertvanduursen - The Guru Of Vision- Fabrizio Pisani- Alexander Hartvig Nielsen- Peggy Youell- Konstantin Shabashov- The Dodd- DGJono - Matthias Meger- Scott Stevens- Emilio Alvarezpatreon.com/robertskmiles
Respectability Robert Miles AI Safety 2017-05-27 | It can be hard to get people to take AI Safety concerns seriously, but it's a lot easier now than it used to be.That Eliezer Yudkowsky talk (containing (a very small amount of) screaming): youtube.com/watch?v=EUjc1WuyPT8The Open Letter: futureoflife.org/ai-open-letterWith thanks to my wonderful Patreon Supporters:- Ichiro Dohi- Stefan Skiles- Chad Jones- Joshua Richardson- Fabian Consiglio- Jonatan R- Øystein Flygt- Björn Mosten- Michael Greve- robertvanduursen - The Guru Of Vision- Fabrizio Pisani- Peggy Youell- Konstantin Shabashov- Adam Dodd- DGJono - Matthias Megerhttp://patreon.com/robertskmiles
Predicting AI: RIP Prof. Hubert Dreyfus Robert Miles AI Safety 2017-05-18 | It's hard to predict what AI will be like in the future. Many tried in the past, and all failed to some extent. In this video we look at Professor Hubert Dreyfus, and one of his reasons for thinking AI couldn't be done.Some of Dreyfus' work:"What Computers Can't Do": archive.org/details/whatcomputerscan017504mbp"Alchemy and Artificial Intelligence": https://courses.csail.mit.edu/6.803/pdf/dreyfussummaryandconclusion.pdfHere's that paper criticising him:"The Artificial Intelligence of Hubert L. Dreyfus: A Budget of Fallacies": https://dspace.mit.edu/handle/1721.1/6084With thanks to my excellent Patreon Supporters:- Ichiro Dohi- Chad Jones- Joshua Richardson- Fabian Consiglio- Jonatan R- Øystein Flygt- Björn Mosten- Peggy Youell- Konstantin Shabashov- Almighty Dodd- DGJonopatreon.com/robertskmiles
Whats the Use of Utility Functions? Robert Miles AI Safety 2017-04-27 | A lot of our problems with AI seem to relate to its utility function. Why do we need one of those, anyway?Footage from The Simpsons, Copyright FOX, used under Fair UseWith thanks to everyone who gave their suggestions and feedback about drafts of this videoAnd thanks to my Patreon Supporters:Chad JonesPeggy YouellSylvain Chevalierpatreon.com/robertskmiles
Where do we go now? Robert Miles AI Safety 2017-03-31 | I wanted to put 'cards' for all the videos as they come up, but I'm only allowed five cards, so just use this playlist of all my Computerphile videos:youtube.com/playlist?list=PLqL14ZxTTA4fRMts7Af2G8t4Rp17e8MdSSee also:Dr Sean Holden's Computerphile Video:youtube.com/watch?v=uA9mxq3gneEBrais Martinez's AlphaGo video:youtube.com/watch?v=qWcfiPi9gUUThanks to Computerphile for permission to use the clips. Just about every time I say "I made" in this video, I mean "Computerphile made, using footage of me".youtube.com/user/ComputerphileSupport me on Patreon, if you like!http://patreon.com/robertskmilesI just set it up, cause people asked for it, and I'm not sure how that all works, but let's figure it out together :)Let me know what you want to see on this channel!
Status Report Robert Miles AI Safety 2017-03-18 | Videos on the way as soon as I can get a computer than can edit videos
Channel Introduction Robert Miles AI Safety 2017-02-28 | Welcome to the channel!I'm new to making YouTube videos myself, so sorry for the bad editing.
