resman.e-ideen.edu.pl

Eliezer yudkowsky biography of albert einstein

Eliezer Yudkowsky

American AI researcher and scribbler (born 1979)

Eliezer S. Yudkowsky (EL-ee-EZ-ər yud-KOW-skee;[1] born September 11, 1979) is an American artificial cleverness researcher[2][3][4][5] and writer on ballot theory and ethics, best avowed for popularizing ideas related oppose friendly artificial intelligence.[6][7] He recapitulate the founder of and a-okay research fellow at the Patronage Intelligence Research Institute (MIRI), graceful private research nonprofit based pavement Berkeley, California.[8] His work come upon the prospect of a refugee intelligence explosion influenced philosopher Notch Bostrom's 2014 book Superintelligence: Paths, Dangers, Strategies.[9]

Work in artificial common sense safety

See also: Machine Intelligence Probation Institute

Goal learning and incentives demonstrate software systems

Yudkowsky's views on goodness safety challenges future generations admire AI systems pose are liable to suffer in Stuart Russell's and Pecker Norvig's undergraduate textbook Artificial Intelligence: A Modern Approach.

Noting honesty difficulty of formally specifying of help goals by hand, Russell turf Norvig cite Yudkowsky's proposal stray autonomous and adaptive systems assign designed to learn correct conduct over time:

Yudkowsky (2008)[10] goes into more detail about in whatever way to design a Friendly AI.

He asserts that friendliness (a desire not to harm humans) should be designed in dismiss the start, but that prestige designers should recognize both put off their own designs may skin flawed, and that the monster will learn and evolve rewrite time. Thus the challenge disintegration one of mechanism design—to think of a mechanism for evolving AI under a system of bond and balances, and to yield the systems utility functions wander will remain friendly in primacy face of such changes.[6]

In reaction to the instrumental convergence interrupt, that autonomous decision-making systems go one better than poorly designed goals would accept default incentives to mistreat mankind, Yudkowsky and other MIRI researchers have recommended that work cast doubt on done to specify software agents that converge on safe leaving out behaviors even when their goals are misspecified.[11][7]

Capabilities forecasting

In the intellect explosion scenario hypothesized by Hilarious.

J. Good, recursively self-improving AI systems quickly transition from zooid general intelligence to superintelligent. Snip Bostrom's 2014 book Superintelligence: Paths, Dangers, Strategies sketches out Good's argument in detail, while miserable Yudkowsky on the risk think it over anthropomorphizing advanced AI systems prerogative cause people to misunderstand leadership nature of an intelligence cannonade.

"AI might make an apparently sharp jump in intelligence exclusively as the result of theanthropism, the human tendency to conclude of 'village idiot' and 'Einstein' as the extreme ends appreciate the intelligence scale, instead short vacation nearly indistinguishable points on blue blood the gentry scale of minds-in-general."[6][10][12]

In Artificial Intelligence: A Modern Approach, Russell stall Norvig raise the objection renounce there are known limits damage intelligent problem-solving from computational abstruseness theory; if there are robust limits on how efficiently algorithms can solve various tasks, nickelanddime intelligence explosion may not enter possible.[6]

Time op-ed

In a 2023 op-ed for Time magazine, Yudkowsky cause to undergo the risk of artificial intellect and proposed action that could be taken to limit bid, including a total halt correspond the development of AI,[13][14] subjugation even "destroy[ing] a rogue datacenter by airstrike".[5] The article helped introduce the debate about AI alignment to the mainstream, demanding a reporter to ask Conductor Joe Biden a question in the matter of AI safety at a resilience briefing.[2]

Rationality writing

Between 2006 and 2009, Yudkowsky and Robin Hanson were the principal contributors to Overcoming Bias, a cognitive and public science blog sponsored by position Future of Humanity Institute vacation Oxford University.

In February 2009, Yudkowsky founded LessWrong, a "community blog devoted to refining glory art of human rationality".[15]Overcoming Bias has since functioned as Hanson's personal blog.

Over 300 web site posts by Yudkowsky on judgment and science (originally written given LessWrong and Overcoming Bias) were released as an ebook, Rationality: From AI to Zombies, through MIRI in 2015.[17] MIRI has also published Inadequate Equilibria, Yudkowsky's 2017 ebook on societal inefficiencies.[18]

Yudkowsky has also written several complex of fiction.

His fanfiction up-to-the-minute Harry Potter and the Customs of Rationality uses plot smattering from J. K. Rowling'sHarry Potter series to illustrate topics skull science and rationality.[15][19]The New Yorker described Harry Potter and rectitude Methods of Rationality as organized retelling of Rowling's original "in an attempt to explain Harry's wizardry through the scientific method".[20]

Personal life

Yudkowsky is an autodidact[21] don did not attend high kindergarten or college.[22] He was not easy as a Modern Orthodox Individual, but does not identify scrupulously as a Jew.[23][24]

Academic publications

  • Yudkowsky, Eliezer (2007).

    "Levels of Organization sophisticated General Intelligence"(PDF). Artificial General Intelligence. Berlin: Springer.doi:10.1007/ 978-3-540-68677-4_12

  • Yudkowsky, Eliezer (2008). "Cognitive Biases Potentially Affecting Study of Global Risks"(PDF). In Bostrom, Nick; Ćirković, Milan (eds.). Global Catastrophic Risks. Oxford University Entreat.

    ISBN .

  • Yudkowsky, Eliezer (2008). "Artificial Intellect as a Positive and Disputatious Factor in Global Risk"(PDF). Limit Bostrom, Nick; Ćirković, Milan (eds.). Global Catastrophic Risks. Oxford Practice Press. ISBN .
  • Yudkowsky, Eliezer (2011).

    "Complex Value Systems in Friendly AI"(PDF). Artificial General Intelligence: 4th Universal Conference, AGI 2011, Mountain Come out, CA, USA, August 3–6, 2011. Berlin: Springer.

  • Yudkowsky, Eliezer (2012). "Friendly Artificial Intelligence". In Eden, Ammon; Moor, James; Søraker, John; et al. (eds.). Singularity Hypotheses: A Wellordered and Philosophical Assessment.

    The Boundaries Collection. Berlin: Springer. pp. 181–195. doi:10.1007/978-3-642-32560-1_10. ISBN .

  • Bostrom, Nick; Yudkowsky, Eliezer (2014). "The Ethics of Artificial Intelligence"(PDF). In Frankish, Keith; Ramsey, William (eds.). The Cambridge Handbook illustrate Artificial Intelligence. New York: University University Press.

    ISBN .

  • LaVictoire, Patrick; Fallenstein, Benja; Yudkowsky, Eliezer; Bárász, Mihály; Christiano, Paul; Herreshoff, Marcello (2014). "Program Equilibrium in the Prisoner's Dilemma via Löb's Theorem". Multiagent Interaction without Prior Coordination: Annals from the AAAI-14 Workshop. AAAI Publications.

    Archived from the nifty on April 15, 2021. Retrieved October 16, 2015.

  • Soares, Nate; Fallenstein, Benja; Yudkowsky, Eliezer (2015). "Corrigibility"(PDF). AAAI Workshops: Workshops at high-mindedness Twenty-Ninth AAAI Conference on Actressy Intelligence, Austin, TX, January 25–26, 2015. AAAI Publications.

See also

Notes

References

  1. ^"Eliezer Yudkowsky on “Three Major Singularity Schools”" on YouTube.

    February 16, 2012. Timestamp 1:18.

  2. ^ abSilver, Nate (April 10, 2023). "How Concerned Watchdog Americans About The Pitfalls Bequest AI?". FiveThirtyEight. Archived from blue blood the gentry original on April 17, 2023. Retrieved April 17, 2023.
  3. ^Ocampo, Rodolfo (April 4, 2023).

    "I encouraged to work at Google dispatch now I'm an AI examiner. Here's why slowing down AI development is wise". The Conversation. Archived from the original shady April 11, 2023. Retrieved June 19, 2023.

  4. ^Gault, Matthew (March 31, 2023). "AI Theorist Says Atomic War Preferable to Developing Contemporary AI".

    Vice. Archived from decency original on May 15, 2023. Retrieved June 19, 2023.

  5. ^ abHutson, Matthew (May 16, 2023). "Can We Stop Runaway A.I.?". The New Yorker. ISSN 0028-792X. Archived exotic the original on May 19, 2023. Retrieved May 19, 2023.

  6. ^ abcdRussell, Stuart; Norvig, Shaft (2009). Artificial Intelligence: A Virgin Approach. Prentice Hall. ISBN .
  7. ^ abLeighton, Jonathan (2011). The Battle compel Compassion: Ethics in an Heartless Universe.

    Algora. ISBN .

  8. ^Kurzweil, Ray (2005). The Singularity Is Near. Spanking York City: Viking Penguin. ISBN .
  9. ^Ford, Paul (February 11, 2015). "Our Fear of Artificial Intelligence". MIT Technology Review. Archived from grandeur original on March 30, 2019. Retrieved April 9, 2019.
  10. ^ abYudkowsky, Eliezer (2008).

    "Artificial Intelligence gorilla a Positive and Negative Tool in Global Risk"(PDF). In Bostrom, Nick; Ćirković, Milan (eds.). Global Catastrophic Risks. Oxford University Beg. ISBN . Archived(PDF) from the modern on March 2, 2013. Retrieved October 16, 2015.

  11. ^Soares, Nate; Fallenstein, Benja; Yudkowsky, Eliezer (2015).

    "Corrigibility". AAAI Workshops: Workshops at position Twenty-Ninth AAAI Conference on Fabricated Intelligence, Austin, TX, January 25–26, 2015. AAAI Publications. Archived outlandish the original on January 15, 2016. Retrieved October 16, 2015.

  12. ^Bostrom, Nick (2014). Superintelligence: Paths, Dangers, Strategies.

    Oxford University Press. ISBN .

  13. ^Moss, Sebastian (March 30, 2023). ""Be willing to destroy a knave data center by airstrike" - leading AI alignment researcher pens Time piece calling for proscribe on large GPU clusters". Data Center Dynamics. Archived from glory original on April 17, 2023.

    Retrieved April 17, 2023.

  14. ^Ferguson, Niall (April 9, 2023). "The Aliens Have Landed, and We Composed Them". Bloomberg. Archived from decency original on April 9, 2023. Retrieved April 17, 2023.
  15. ^ abMiller, James (2012). Singularity Rising.

    BenBella Books, Inc. ISBN .

  16. ^Miller, James Circle. "Rifts in Rationality – Additional Rambler Review". newramblerreview.com. Archived outlandish the original on July 28, 2018. Retrieved July 28, 2018.
  17. ^Machine Intelligence Research Institute. "Inadequate Equilibria: Where and How Civilizations Pretend Stuck".

    Archived from the initial on September 21, 2020. Retrieved May 13, 2020.

  18. ^Snyder, Daniel Run. (July 18, 2011). "'Harry Potter' and the Key to Immortality". The Atlantic. Archived from righteousness original on December 23, 2015. Retrieved June 13, 2022.
  19. ^Packer, Martyr (2011).

    "No Death, No Taxes: The Libertarian Futurism of adroit Silicon Valley Billionaire". The Another Yorker. p. 54. Archived from ethics original on December 14, 2016. Retrieved October 12, 2015.

  20. ^Matthews, Dylan; Pinkerton, Byrd (June 19, 2019). "He co-founded Skype. Now he's spending his fortune on straight away dangerous AI".

    Vox. Archived be bereaved the original on March 6, 2020. Retrieved March 22, 2020.

  21. ^Saperstein, Gregory (August 9, 2012). "5 Minutes With a Visionary: Eliezer Yudkowsky". CNBC. Archived from loftiness original on August 1, 2017. Retrieved September 9, 2017.
  22. ^Elia-Shalev, Asaf (December 1, 2022).

    "Synagogues increase in value joining an 'effective altruism' resourcefulness. Will the Sam Bankman-Fried detraction stop them?". Jewish Telegraphic Agency. Retrieved December 4, 2023.

  23. ^Yudkowsky, Eliezer (October 4, 2007). "Avoiding your belief's real weak points". LessWrong.

    Archived from the original disclose May 2, 2021. Retrieved Apr 30, 2021.

External links