Eric Horvitz

Horvitz received his Ph.D and M.D. from Stanford University.{{Cite web|url=https://ieeexplore.ieee.org/author/37282933300|title=Eric Horvitz|website=IEEE Xplore Digital Library|access-date=3 June 2019}} His doctoral dissertation, Computation and Action Under Bounded Resources,{{Cite web |last=Horvitz |first=Eric |date=1990 |title=Computation and action under bounded resources |url=https://erichorvitz.com/horvitzdiss.pdf}} and follow-on research introduced models of bounded rationality founded in probability and decision theory. He did his doctoral work under advisors Ronald A. Howard, George B. Dantzig, Edward H. Shortliffe, and Patrick Suppes.

He is currently the Chief Scientific Officer of Microsoft. He has been elected Fellow of the Association for the Advancement of Artificial Intelligence (AAAI), the National Academy of Engineering (NAE), the American Academy of Arts and Sciences, and of the American Association for the Advancement of Science (AAAS).

He was elected as an ACM Fellow in 2014 for "contributions to artificial intelligence, and human-computer interaction."[http://awards.acm.org/award_winners/horvitz_4722252.cfm ERIC HORVITZ ACM Fellows 2014]

He was elected to the ACM CHI Academy in 2013 for “research at the intersection of human-computer interaction and artificial intelligence.”{{Cite web |title=2013 SIGCHI Awards – ACM SIGCHI |url=https://sigchi.org/awards/sigchi-award-recipients/2013-sigchi-awards/ |access-date=2022-03-22 |language=en-US}}

He was elected to the American Philosophical Society in 2018.{{Cite web|url=https://www.amphilsoc.org/blog/election-new-members-2018-spring-meeting|title=Election of New Members at the 2018 Spring Meeting | American Philosophical Society}}

In 2015, he was awarded the AAAI Feigenbaum Prize,{{cite web|title=The AAAI Feigenbaum Prize|url=http://www.aaai.org/Awards/feigenbaum.php|publisher=AAAI|access-date=14 April 2016}} a biennial award for sustained and high-impact contributions to the field of artificial intelligence through the development of computational models of perception, reflection and action, and their application in time-critical decision making, and intelligent information, traffic, and healthcare systems.

In 2015, he was also awarded the ACM - AAAI Allen Newell Award,{{cite web|url=http://awards.acm.org/award_winners/horvitz_4722252.cfm|title=ERIC HORVITZ - Award Winner|publisher=ACM|access-date=3 June 2019}} for "contributions to artificial intelligence and human-computer interaction spanning the computing and decision sciences through developing principles and models of sensing, reflection, and rational action."

He serves on the President's Council of Advisors on Science and Technology{{Cite web|title=Eric Horvitz, MD PhD|url=https://bidenwhitehouse.archives.gov/pcast/members/eric-horvitz/|access-date=2022-01-05|website=The White House|language=en-US}}, the Scientific Advisory Committee of the Allen Institute for Artificial Intelligence (AI2), and the [https://www.nationalacademies.org/cstb/about Computer Science and Telecommunications Board] (CSTB) {{Cite web|title=Membership of the Computer Science and Telecommunications Board|url=http://sites.nationalacademies.org/CSTB/CSTB_042202|access-date=3 June 2019|website=The National Academies of Sciences, Engineering and Medicine|archive-date=3 June 2019|archive-url=https://web.archive.org/web/20190603173942/http://sites.nationalacademies.org/CSTB/CSTB_042202|url-status=dead}} (PCAST), the Scientific Advisory Committee of the Allen Institute for Artificial Intelligence (AI2), and the Computer Science and Telecommunications Board{{Cite web|title=About the Computer Science and Telecommunications Board|url=https://www.nationalacademies.org/cstb/about#sl-three-columns-d81de41d-2831-48f0-8623-02c1b399f6d5|access-date=2022-01-14|website=National Academies}} (CSTB) of the US National Academies.

He has served as president of the Association for the Advancement of AI (AAAI), on the NSF Computer & Information Science & Engineering (CISE) Advisory Board, on the council of the Computing Community Consortium (CCC), chair of the Section on Information, Computing, and Communications of the American Association for the Advancement of Science (AAAS), on the Board of Regents{{Cite web|title=Board of Regents|url=https://www.nlm.nih.gov/od/bor/bor.html|access-date=3 June 2019|website=U.S. National Library of Medicine}} of the US National Library of Medicine (NLM), and a member of the National Security Commission on Artificial Intelligence (NSCAI) that was established in 2018 and issued its final report in March 2021.{{Cite news |last=Simonite |first=Tom |title=This Group Pushed More AI in US Security—and Boosted Big Tech |url=https://www.wired.com/story/group-pushed-ai-us-security-boosted-tech/ |access-date=2025-02-04 |work=Wired |language=en-US |issn=1059-1028}}{{Cite web |last=Shead |first=Sam |date=2021-03-02 |title=U.S. is ‘not prepared to defend or compete in the A.I. era,’ says expert group chaired by Eric Schmidt |url=https://www.cnbc.com/2021/03/02/us-not-prepared-to-defend-or-compete-in-ai-era-says-eric-schmidt-group.html |access-date=2025-02-04 |website=CNBC |language=en}}

Horvitz's research interests span theoretical and practical challenges with developing systems that perceive, learn, and reason. His contributions include advances in principles and applications of machine learning and inference, information retrieval, human-computer interaction, bioinformatics, and e-commerce.

Horvitz played a significant role in the use of probability and decision theory in artificial intelligence. His work raised the credibility of artificial intelligence in other areas of computer science and computer engineering, influencing fields ranging from human-computer interaction to operating systems. His research helped establish the link between artificial intelligence and decision science. As an example, he coined the concept of bounded optimality, a decision-theoretic approach to bounded rationality.{{cite news|last=Mackworth|first=Alan|title=Introduction of Eric Horvitz|url=http://research.microsoft.com/en-us/um/people/horvitz/AAAI_Presidential%20Address_Eric_Horvitz.pdf|newspaper=AAAI Presidential Address|date=July 2008}} The influences of bounded optimality extend beyond computer science into cognitive science and psychology.{{Cite journal|last1=Gershman|first1=Samuel J.|last2=Horvitz|first2=Eric J.|last3=Tenenbaum|first3=Joshua B.|date=17 July 2015|title=Computational rationality: A converging paradigm for intelligence in brains, minds, and machines|journal=Science|volume=349|issue=6245|pages=273–278|doi=10.1126/science.aac6076|pmid=26185246|bibcode=2015Sci...349..273G|s2cid=14818619}}{{Cite journal|last1=Howes|first1=Andrew|last2=Duggan|first2=Geoffrey B.|last3=Kalidindi|first3=Kiran|last4=Tseng|first4=Yuan-Chi|last5=Lewis|first5=Richard L.|date=1 July 2016|title=Predicting Short-term Remembering as Boundedly Optimal Strategy Choice|url=https://research.birmingham.ac.uk/portal/files/23709347/Howes_2015_EmailCal.pdf|journal=Cognitive Science|volume=40|issue=5|pages=1192–1223|doi=10.1111/cogs.12271|pmid=26294328|doi-access=free}}

He studied the use of probability and utility to guide automated reasoning for decision making. The methods include consideration of the solving of streams of problems{{Citation|last=Horvitz|first=Eric|title=Principles and Applications of Continual Computation|url=http://research.microsoft.com/~horvitz/ccprinciples.htm|journal=Artificial Intelligence|volume=126|issue=1–2|pages=159–196|date=February 2001|doi=10.1016/S0004-3702(00)00082-5|citeseerx=10.1.1.476.5653}} in environments over time. In related work, he applied probability and machine learning to solve combinatorial problems and to guide theorem proving.{{Citation|last1=Horvitz|first1=Eric J.|last2=Ruan|first2=Y.|last3=Gomes|first3=C.|last4=Kautz|first4=H.|last5=Selman|first5=B.|last6=Chickering|first6=D.M.|title=A Bayesian Approach to Tackling Hard Computational Problems|url=http://research.microsoft.com/en-us/um/people/horvitz/UAI2k1.pdf |journal=Proceedings of the Conference on Uncertainty and Artificial Intelligence|pages=235–244|date=July 2001}} He introduced the anytime algorithm paradigm in AI,{{Cite journal|last=Horvitz|first=Eric|date=July 1987|title=Reasoning about beliefs and actions under computational resource constraints|url=http://erichorvitz.com/U87.PDF|journal=UAI'87 Proceedings of the Third Conference on Uncertainty in Artificial Intelligence|publisher=AUAI Press|location=Arlington, VA|pages=429–447|isbn=0-444-87417-8}}{{Cite journal|last=Horvitz|first=Eric|date=August 1988|title=Reasoning under varying and uncertain resource constraints|url=http://erichorvitz.com/ftp/AI88.PDF|journal=AAAI'88 Proceedings of the Seventh AAAI National Conference on Artificial Intelligence|publisher=AAAI Press|pages=111–116}} where partial results, probabilities, or utilities of outcomes are refined with computation under different availabilities or costs of time, guided by the expected value of computation.{{Cite journal|last1=Horvitz|first1=Eric J.|last2=Cooper|first2=Gregory F.|last3=Heckerman|first3=David E.|date=August 1989|title=Reflection and action under scarce resources: theoretical principles and empirical study|url=http://erichorvitz.com/ij89.pdf|journal=IJCAI'89 Proceedings of the 11th International Joint Conference on Artificial Intelligence - Volume 2|publisher=Morgan Kaufmann Publishers Inc.|location=San Francisco, CA|pages=1121–1127}}{{Cite thesis |degree=PhD |last=Horvitz|first=Eric|date=December 1990|title=Computation and Action Under Bounded Resources|url=http://erichorvitz.com/horvitzdiss.pdf |publisher=Stanford University}}

He has issued long-term challenge problems for AI{{Citation|last1=Selman|first1=B.|last2=Brooks|first2=R.|last3=Dean|first3=T.|last4=Horvitz|first4=E.|last5=Mitchell|first5=T.|last6=Nilsson|first6=N.|title=Challenge Problems for Artificial Intelligence|url=http://www.research.microsoft.com/~horvitz/selman.htm|journal=Proceedings of AAAI-96, Thirteenth National Conference on Artificial Intelligence, Portland, Oregon|pages=1340–1345|date=August 1996}}—and has espoused a vision of open-world AI,{{Citation|last=Horvitz|first=Eric|title=Artificial Intelligence in the Open World|url=http://research.microsoft.com/en-us/um/people/horvitz/AAAI_Presidential_Lecture_Eric_Horvitz.htm|journal=AAAI Presidential Lecture|date=July 2008}} where machine intelligences have the ability to understand and perform well in the larger world where they encounter situations they have not seen before.

He has explored synergies between human and machine intelligence.{{Cite web |last=Horvitz |first=Eric |date=Jan 13, 2020 |title=2019 Annual Meeting Plenary: People, Machines, and Intelligence |url=https://www.youtube.com/watch?v=cu_609pnVDE |access-date=2020-04-02 |website=YouTube}} He introduced principles for using machine learning and decision theory to guide machine versus human initiative,{{Cite book |last=Horvitz |first=Eric |title=Proceedings of the SIGCHI conference on Human factors in computing systems the CHI is the limit - CHI '99 |chapter=Principles of mixed-initiative user interfaces |date=May 1999 |chapter-url=http://erichorvitz.com/chi99horvitz.pdf |location=New York, NY |publisher=ACM |pages=159–166 |citeseerx=10.1.1.78.7280 |doi=10.1145/302979.303030 |isbn=0-201-48559-1 |s2cid=8943607}} methods that provide AI systems with understandings of when to transfer problem solving to humans,{{Cite journal |last1=Horvitz |first1=Eric |last2=Paek |first2=Tim |date=2007-03-01 |title=Complementary computing: policies for transferring callers from dialog systems to human receptionists |url=https://doi.org/10.1007/s11257-006-9026-1 |journal=User Modeling and User-Adapted Interaction |language=en |volume=17 |issue=1 |pages=159–182 |doi=10.1007/s11257-006-9026-1 |s2cid=2819237 |issn=1573-1391|url-access=subscription }} and the use of machine learning and planning techniques to identify and merge the complementary abilities of people and AI systems.{{Cite journal |last1=Kamar |first1=Ece |last2=Hacker |first2=Severin |last3=Horvitz |first3=Eric |date=8 June 2018 |title=Combining human and machine intelligence in large-scale crowdsourcing |url=http://erichorvitz.com/CrowdSynth.pdf |journal=Proceeding, AAMAS '12 Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems - Volume 1 |location=Richland, SC |publisher=International Foundation for Autonomous Agents and Multiagent Systems |volume=1 |pages=467–474 |isbn=978-0-9817381-1-6}}{{Cite book |last1=Wilder |first1=Bryan |last2=Horvitz |first2=Eric |last3=Kamar |first3=Ece |title=Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence |chapter=Learning to Complement Humans |date=2020-07-09 |chapter-url=https://www.ijcai.org/proceedings/2020/212 |language=en |volume= 2|pages=1526–1533 |doi=10.24963/ijcai.2020/212|arxiv=2005.00582 |isbn=978-0-9992411-6-5 |s2cid=218486980 }} In work on human-centered AI, he introduced measures and models of the expected value of displayed information to guide the display of information to human decision makers in time-critical settings{{Cite journal |last1=Horvitz |first1=Eric |last2=Barry |first2=Matthew |date=August 1995 |title=Display of information for time-critical decision making |url=http://erichorvitz.com/cognitive_cost_display_of_information_UAI_95.pdf |journal=Proceeding, UAI'95 Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence |location=San Francisco, CA |publisher=Morgan Kaufmann Publishers Inc |pages=296–305 |isbn=1-55860-385-9}} and methods for making statistical AI inferences more understandable.{{Cite journal |last1=Horvitz |first1=Eric |last2=Heckerman |first2=David |last3=Nathwani |first3=Bharat |last4=Fagan |first4=Lawrence |date=October 1986 |title=The use of a heuristic problem-solving hierarchy to facilitate the explanation of hypothesis-directed reasoning |url=https://erichorvitz.com/medinfo_explain_inference.pdf |journal=Proceedings of Medinfo |pages=27–31}} He introduced models of human attention in computing systems,{{Cite journal |last1=Horvitz |first1=Eric |last2=Kadie |first2=Carl |last3=Peak |first3=Tim |last4=Hovel |first4=David |date=March 2003 |title=Models of attention in computing and communication: from principles to applications |url=http://erichorvitz.com/Models_of_attention_in_computing.pdf |journal=Communications of the ACM |location=New York, NY |publisher=ACM |volume=46 |pages=52–59 |doi=10.1145/636772.636798 |s2cid=2584780}}{{Cite news |last=Markhoff |first=John |date=17 July 2000 |title=Microsoft Sees Software 'Agent' as Way to Avoid Distractions |work=The New York Times |url=https://archive.nytimes.com/www.nytimes.com/library/tech/00/07/biztech/articles/17lab.html |access-date=3 June 2019}} and studied the use of machine learning to infer the cost of interruptions to computer users.{{Cite book |last1=Horvitz |first1=Eric |last2=Apacible |first2=Johnson |title=Proceedings of the 5th international conference on Multimodal interfaces |chapter=Learning and reasoning about interruption |date=2003-11-05 |chapter-url=https://doi.org/10.1145/958432.958440 |series=ICMI '03 |location=New York, NY, USA |publisher=Association for Computing Machinery |pages=20–27 |doi=10.1145/958432.958440 |isbn=978-1-58113-621-0|s2cid=1183716 }}{{Cite book |last1=Horvitz |first1=Eric |last2=Koch |first2=Paul |last3=Apacible |first3=Johnson |title=Proceedings of the 2004 ACM conference on Computer supported cooperative work |chapter=BusyBody |date=2004-11-06 |chapter-url=https://doi.org/10.1145/1031607.1031690 |series=CSCW '04 |location=New York, NY, USA |publisher=Association for Computing Machinery |pages=507–510 |doi=10.1145/1031607.1031690 |isbn=978-1-58113-810-8|s2cid=11517148 }} His use of machine learning to build models of human surprise was featured as a technology breakthrough by MIT Technology Review.{{Cite web |last=Waldrop |first=M. Mitchell |date=19 February 2008 |title=TR10: Modeling Surprise |url=http://www2.technologyreview.com/news/409594/tr10-modeling-surprise/ |access-date=2022-01-11 |website=MIT Technology Review |language=en-us}}

He investigated the use of AI methods to provide assistance to users including help with software{{Cite journal|last1=Horvitz|first1=E|last2=Breese|first2=J|last3=Heckerman|first3=D|last4=Hovel|first4=D|last5=Rommelse|first5=K|date=July 1998|title=The Lumiere Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users|url=http://erichorvitz.com/lumiere.htm|journal=Proceedings of UAI, Conference on Uncertainty in Artificial Intelligence|pages=256–265}}{{Cite web|url=https://www.youtube.com/watch?v=bt-JXQS0zYc|title=Lumiere - Intelligent User Interface|date=2009-08-31|website=YouTube|orig-date=1995|access-date=2020-04-02}} and in the daily life.{{Cite web|date=2017-02-09|title=How the forgetfulness of one of Microsoft's top scientists inspired a killer new feature for Windows 10|url=https://www.businessinsider.com/microsoft-research-eric-horvitz-cortana-commitments-2017-2|access-date=2022-01-11|website=Business Insider Australia|language=en-AU}}

He made contributions to multimodal interaction.{{Cite web|url=https://www.youtube.com/watch?v=1KYxJMU8ks4|title=Information Agents: Directions and Futures (2001)|date=2020-02-21|website=YouTube|orig-date=2001|access-date=2020-04-02}} In 2015, he received the ACM ICMI Sustained Accomplishment Award{{Cite web|url=https://icmi.acm.org/2015/index.php?id=award|title=ICMI Sustained Accomplishment Award|website=International Conference on Multimodal Interaction|access-date=2020-04-02}}{{Cite journal|last1=Bohus|first1=D|last2=Horvitz|first2=E|date=November 2009|title=Dialog in the Open World: Platform and Applications|url=http://erichorvitz.com/icmi2009_bohus_horvitz.pdf|journal=ACM International Conference on Multimodal Interaction}} for contributions to multimodal interaction. His work on multimodal interaction includes studies of situated interaction,{{Cite book|last1=Bohus|first1=D|title=The Handbook of Multimodal-Multisensor Interfaces, Volume 3|last2=Horvitz|first2=E|publisher=Association for Computing Machinery and Morgan & Claypool|year=2019|isbn=978-1-970001-75-4|pages=105–143|chapter=Situated Interaction}}{{Cite web|url=https://www.youtube.com/watch?v=IjBNy8nR24E|title=Elevating human-computer interaction to a new level of sophistication|date=2014-04-08|website=YouTube|access-date=2020-04-02}} where systems consider physical details of open-world settings and can perform dialog with multiple people.{{Cite book|last1=Bohus|first1=D|last2=Horvitz|first2=E|title=Proceedings of the SIGDIAL 2009 Conference on the 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue - SIGDIAL '09|chapter=Models for multiparty engagement in open-world dialog|date=September 2009|chapter-url=http://erichorvitz.com/Bohus_Horvitz_Multiparty_Engagement.pdf|pages=225–234|doi=10.3115/1708376.1708409|isbn=9781932432640|s2cid=2125746}}

He co-authored probability-based methods to enhance privacy, including a model of altruistic sharing of data called community sensing{{Citation|last1=Krause|first1=A.|last2=Horvitz|first2=E.|last3=Kansal|first3=A.|last4=Zhao|first4=F.|chapter=Toward Community Sensing|chapter-url=http://research.microsoft.com/en-us/um/people/horvitz/Community_Sensing.htm|title=Proceedings of IPSN 2008|date=April 2008}} and risk-sensitive approaches{{Cite journal|last1=Krause|first1=Andreas|last2=Horvitz|first2=Eric|date=November 2010|title=A Utility-Theoretic Approach to Privacy in Online Services|url=https://erichorvitz.com/pvoijair_jair.pdf|journal=Journal of Artificial Intelligence Research|volume=39|pages=633–662|doi=10.1613/jair.3089|s2cid=1582192}} including stochastic privacy.{{Citation|last1=Singla|first1=A.|last2=Horvitz|first2=E.|last3=Kamar|first3=E.|last4=White|first4=R.W.|title=Stochastic Privacy|url=http://research.microsoft.com/en-us/um/people/horvitz/stochastic_privacy.pdf|journal=AAAI|date=July 2014|volume=28 |doi=10.1609/aaai.v28i1.8734 |bibcode=2014arXiv1404.5454S|arxiv=1404.5454|s2cid=14584347 }}

He is Microsoft's top inventor.{{Cite web|date=2019-05-28|title=Swimming in Creative Waters: The Art of Invention|url=https://www.microsoft.com/en-us/research/blog/swimming-in-creative-waters-a-young-artist-an-inventor-and-a-nurturing-sea-of-family-and-colleagues/|access-date=2022-01-13|website=Microsoft Research|language=en-US}}

He led efforts in applying AI methods to computing systems, including machine learning for memory management in Windows,{{Cite web|last=Keizer|first=Gregg|date=2007-01-19|title=Microsoft Predicts The Future With Vista's SuperFetch|url=https://www.informationweek.com/software/microsoft-predicts-the-future-with-vista-s-superfetch|access-date=2022-01-13|website=InformationWeek|language=en}} web prefetching,{{Cite book|last=Horvitz|first=Eric|title=Proceedings of the seventh international conference on Information and knowledge management |chapter=Continual computation policies for utility-directed prefetching |date=1998-11-01|chapter-url=https://doi.org/10.1145/288627.288655|series=CIKM '98|location=New York, NY, USA|publisher=Association for Computing Machinery|pages=175–184|doi=10.1145/288627.288655|isbn=978-1-58113-061-4|s2cid=13518550}} graphics rendering,{{cite arXiv|last1=Horvitz|first1=Eric J.|last2=Lengyel|first2=Jed|date=2013-02-06|title=Perception, Attention, and Resources: A Decision-Theoretic Approach to Graphics Rendering|class=cs.AI|eprint=1302.1547}} and web crawling.{{Cite book|last1=Kolobov|first1=Andrey|last2=Peres|first2=Yuval|last3=Lubetzky|first3=Eyal|last4=Horvitz|first4=Eric|title=Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval |chapter=Optimal Freshness Crawl Under Politeness Constraints |date=2019-07-18|chapter-url=https://doi.org/10.1145/3331184.3331241|series=SIGIR'19|location=New York, NY, USA|publisher=Association for Computing Machinery|pages=495–504|doi=10.1145/3331184.3331241|isbn=978-1-4503-6172-9|s2cid=196203203}} He did early work on AI for debugging software.{{Cite journal|last1=Burnell|first1=Lisa|last2=Horvitz|first2=Eric|date=1995-03-01|title=Structure and chance: melding logic and probability for software debugging|url=https://doi.org/10.1145/203330.203338|journal=Communications of the ACM|volume=38|issue=3|pages=31–ff|doi=10.1145/203330.203338|s2cid=9644536|issn=0001-0782}}

Horvitz speaks on the topic of artificial intelligence, including on NPR and the Charlie Rose show.{{cite news|last=Hansen|first=Liane|title=Meet Laura, Your Virtual Personal Assistant|url=https://www.npr.org/templates/story/story.php?storyId=102205784|access-date=16 March 2011|newspaper=NPR|date=21 March 2009}}{{cite web | url=https://www.npr.org/2011/01/11/132840775/The-Singularity-Humanitys-Last-Invention |title=The Singularity: Humanity's Last Invention? | last=Kaste | first=Martin | date=11 Jan 2011 | access-date=14 Feb 2011 | publisher=NPR}}{{cite web|last=Rose|first=Charlie|title=A panel discussion about Artificial Intelligence|url=http://www.charlierose.com/guest/view/1221|access-date=2011-03-12|archive-url=https://web.archive.org/web/20110213142757/http://www.charlierose.com/guest/view/1221|archive-date=2011-02-13}} Online talks include both technical lectures and presentations for general audiences, for example with the TEDx talk Making Friends with Artificial Intelligence.{{Cite AV media |url=https://www.youtube.com/watch?v=dpoVh9xwdD4 |title=Making Friends With Artificial Intelligence: Eric Horvitz at TEDxAustin |date=2013-02-19 |last=TEDx Talks |access-date=2025-06-02 |via=YouTube}} His research has been featured in The New York Times and MIT Technology Review.{{cite news|last=Markoff|first=John|title=Microsoft Introduces Tool for Avoiding Traffic Jams|url=https://www.nytimes.com/2008/04/10/technology/10maps.html?_r=2&adxnnl=1&adxnnlx=1300042441-97RrVwmDAFO2sofP0TWbGg|access-date=16 March 2011|newspaper=The New York Times|date=10 April 2008}}{{cite news|last=Markoff|first=John|title=Microsoft Sees Software 'Agent' as Way to Avoid Distractions|url=https://www.nytimes.com/library/tech/00/07/biztech/articles/17lab.html|access-date=16 March 2011|newspaper=The New York Times|date=17 July 2000}}{{cite news|last=Lohr|first=Steve, and Markoff, John|title=Smarter Than You Think: Computers Learn to Listen, and Some Talk Back|url=https://www.nytimes.com/2010/06/25/science/25voice.html|access-date=12 March 2011|newspaper=The New York Times|date=24 June 2010}}

He has testified before the US Senate on progress, opportunities, and challenges with AI.{{Cite web|url=http://erichorvitz.com/Senate_Testimony_Eric_Horvitz.pdf|title=Reflections on the Status and Future of Artificial Intelligence|last=Horvitz|first=Eric|date=30 November 2016|website=erichorvitz.com|access-date=3 June 2019}}

He has addressed technical and societal challenges and opportunities with the fielding of AI technologies in the open world, including beneficial uses of AI,{{Cite web|url=https://www.youtube.com/watch?v=-uHavVE018M|title=AI for Social Good (2016): Keynote AI in Support of People and Society|date=2016-06-13|website=YouTube|access-date=2020-04-20}} AI safety and robustness,{{Cite web|url=http://erichorvitz.com/OSTP-CMU_AI_Safety_framing_talk.pdf|title=Reflections on Safety and Artificial Intelligence|last=Horvitz|first=Eric|date=2016-06-27|website=Eric Horvitz |access-date=2020-04-20}} and where AI systems and capabilities can have inadvertent effects, pose dangers, or be misused.{{Cite journal|last=Horvitz|first=Eric|date=7 July 2017|title=AI, people, and society|journal=Science|volume=357|issue=6346|pages=7|doi=10.1126/science.aao2466|pmid=28684472|doi-access=free|bibcode=2017Sci...357....7H}}{{Cite journal|last1=Dietterich|first1=Thomas G|last2=Horvitz|first2=Eric J.|date=October 2015|title=Rise of Concerns about AI: Reflections and Directions|url=http://erichorvitz.com/CACM_Oct_2015-VP.pdf|journal=Communications of the ACM|volume=58|issue=10|pages=38–40|doi=10.1145/2770869|doi-access=free}}{{Cite web|url=https://www.youtube.com/watch?v=PjJHPko4J2k|title=Conference on Ethics & AI: Keynote Session|date=2018-04-09|website=YouTube|access-date=2020-04-20}}{{Cite web|url=https://events.technologyreview.com/video/watch/eric-horvitz-microsoft-research-ai-directions-challenges-futures|title=The Long View: AI Directions, Challenges, and Futures|last=Horvitz|first=Eric|date=2017-03-28|website=YouTube|access-date=2020-04-20}} He has presented on caveats with applications of AI in military settings.{{Cite web|url=https://www.youtube.com/watch?v=TUtUTZvZ1-4|title=Keynote Address, Eric Horvitz: AI Advances, Aspirations—and Concerns|date=2019-11-15|website=YouTube|access-date=2020-04-20}} He and Thomas G. Dietterich called for work on AI alignment, saying that AI systems "must reason about what people intend rather than carrying out commands literally."

He has called for action on potential risks to civil liberties posed by government uses of data in AI systems.{{Cite web|last1=Horvitz|first1=Eric|last2=Clyburn|first2=Mignon|last3=Felten|first3=Ed|last4=LeBlanc|first4=Travis|date=2021-05-17|title=Caution ahead: Navigating risks to freedoms posed by AI|url=https://thehill.com/blogs/congress-blog/technology/553932-caution-ahead-navigating-risks-to-freedoms-posed-by-ai|access-date=2022-01-19|website=TheHill|language=en}} He and privacy scholar Deirdre Mulligan stated that society must balance privacy concerns with benefits of data for social benefit.{{Cite news|last1=Horvitz|first1=Eric|last2=Mulligan|first2=Deirdre|date=17 July 2015|title=Data, privacy, and the greater good|volume=349|pages=253–254|work=Science|issue=6245|url=https://erichorvitz.com/data_privacy_greater_good.pdf|access-date=19 Jan 2022}}

He has presented on the risks of AI-enabled deepfakes and contributed to media provenance technologies{{Citation|last1=England|first1=Paul|date=2021-07-15|work=Proceedings of the 12th ACM Multimedia Systems Conference|pages=108–121|place=New York, NY, USA|publisher=Association for Computing Machinery|doi=10.1145/3458305.3459599|isbn=978-1-4503-8434-6|last2=Malvar|first2=Henrique S.|last3=Horvitz|first3=Eric|last4=Stokes|first4=Jack W.|last5=Fournet|first5=Cédric|last6=Burke-Aguero|first6=Rebecca|last7=Chamayou|first7=Amaury|last8=Clebsch|first8=Sylvan|last9=Costa|first9=Manuel|title=AMP: Authentication of media via provenance |s2cid=210859168|doi-access=free}} that cryptographically certify the source and history of edits of digital content.{{Cite web|last=Horvitz|first=Eric|date=2021-02-22|title=A promising step forward on disinformation|url=https://blogs.microsoft.com/on-the-issues/2021/02/22/deepfakes-disinformation-c2pa-origin-cai/|access-date=2022-01-19|website=Microsoft On the Issues|language=en-US}}

He served as President of the AAAI from 2007–2009. As AAAI President, he called together and co-chaired the Asilomar AI study which culminated in a meeting of AI scientists at Asilomar in February 2009. The study considered the nature and timing of AI successes and reviewed concerns about directions with AI developments, including the potential loss of control over computer-based intelligences, and also efforts that could reduce concerns and enhance long-term societal outcomes. The study was the first meeting of AI scientists to address concerns about superintelligence and loss of control of AI and attracted interest by the public.{{cite news|last=Markoff|first=John|title=Scientists Worry Machines May Outsmart Man|url=https://www.nytimes.com/2009/07/26/science/26robot.html| newspaper=York Times|date=26 July 2009}}

In coverage of the Asilomar study, he said that scientists must study and respond to notions of superintelligent machines and concerns about artificial intelligence systems escaping from human control. In a later NPR interview, he said that investments in scientific studies of superintelligences would be valuable to guide proactive efforts even if people believed that the probability of losing of control of AI was low because of the cost of such outcomes.{{cite news|last=Siegel|first=Robert|title=The Singularity: Humanity's Last Invention?|url=https://www.npr.org/2011/01/11/132840775/The-Singularity-Humanitys-Last-Invention| newspaper=NPR|date=11 January 2011}}

In 2014, Horvitz defined and funded with his wife the One Hundred Year Study of Artificial Intelligence (AI100) at Stanford University.{{cite news|last=You|first=Jia|title=A 100-year study of artificial intelligence? Microsoft Research's Eric Horvitz explains|url=http://news.sciencemag.org/people-events/2015/01/100-year-study-artificial-intelligence-microsoft-research-s-eric-horvitz| newspaper=Science|date=9 January 2015}}{{cite news|url=https://www.nytimes.com/2014/12/16/science/century-long-study-will-examine-effects-of-artificial-intelligence.html|title=Study to Examine Effects of Artificial Intelligence|last1=Markoff|first1=John|date=15 December 2014|work=The New York Times|access-date=1 October 2016}} In 2016, the AI Index was launched as a project of the One Hundred Year Study.{{Cite web |date=2016 |title=Preface {{!}} One Hundred Year Study on Artificial Intelligence (AI100) |url=https://ai100.stanford.edu/2016-report/preface |access-date=2025-06-02 |website=ai100.stanford.edu |language=en}}

According to Horvitz, the AI100 gift, which may increase in the future, is sufficient to fund the study for a century. A Stanford press release stated that sets of committees over a century will "study and anticipate how the effects of artificial intelligence will ripple through every aspect of how people work, live and play." A framing memo for the study calls out 18 topics for consideration, including law, ethics, the economy, war, and crime.{{cite web|url=https://stanford.app.box.com/s/266hrhww2l3gjoy9euar|title=One-Hundred Year Study of Artificial Intelligence: Reflections and Framing|date=2014|publisher=Eric Horvitz|access-date=1 October 2016}} Topics include abuses of AI that could pose threats to democracy and freedom and addressing possibilities of superintelligences and loss of control of AI.

The One Hundred Year Study is overseen by a Standing Committee. The Standing Committee formulates questions and themes and organizes a Study Panel every five years. The Study Panel issues a report that assesses the status and rate of progress of AI technologies, challenges, and opportunities with regard to AI's influences on people and society.

The 2015 study panel of the One Hundred Year Study, chaired by Peter Stone, released a report in September 2016, titled "Artificial Intelligence and Life in 2030".{{Cite web |date=2016 |title=Artificial intelligence and life in 2030 |url=https://ai100.stanford.edu/sites/g/files/sbiybj9861/f/ai_100_report_0831fnl.pdf |url-status=dead |archive-url=https://web.archive.org/web/20190530152437/https://ai100.stanford.edu/sites/g/files/sbiybj9861/f/ai_100_report_0831fnl.pdf |archive-date=2019-05-30 |website=ai100.stanford.edu}}{{cite news|url=http://www.chron.com/news/houston-texas/article/New-report-shows-how-artificial-intelligence-will-9195596.php|title=Report: Artificial intelligence to transform urban cities|date=1 September 2016|work=Houston Chronicle|access-date=1 October 2016}} The panel advocated for increased public and private spending on the industry, recommended increased AI expertise at all levels of government, and recommended against blanket government regulation.{{cite news|url=http://www.csmonitor.com/Science/2016/0904/AI-in-the-real-world-Tech-leaders-consider-practical-issues|title=AI in the real world: Tech leaders consider practical issues.|last1=Dussault|first1=Joseph|date=4 September 2016|work=The Christian Science Monitor|access-date=1 October 2016}}Peter Stone et al. "Artificial Intelligence and Life in 2030." One Hundred Year Study on Artificial Intelligence: Report of the 2015-2016 Study Panel, Stanford University, Stanford, CA, September 2016. Doc: http://ai100.stanford.edu/2016-report. Accessed: October 1, 2016. Panel chair Peter Stone argues that AI won't automatically replace human workers, but rather, will supplement the workforce and create new jobs in tech maintenance. While mainly focusing on the next 15 years, the report touched on concerns and expectations that had risen in prominence over the last decade about the risks of superintelligent robots, stating "Unlike in the movies, there's no race of superhuman robots on the horizon or probably even possible.{{cite news|last1=Knight|first1=Will|title=Artificial intelligence wants to be your bro, not your foe|url=https://www.technologyreview.com/s/602278/ai-wants-to-be-your-bro-not-your-foe/|access-date=1 October 2016|work=MIT Technology Review|date=1 September 2016}} Stone stated that "it was a conscious decision not to give credence to this in the report."

The report of the second cycle of the AI100 study,{{Cite web |date=2021 |title=Gathering Strength, Gathering Storms: The One Hundred Year Study on Artificial Intelligence (AI100) 2021 Study Panel Report {{!}} One Hundred Year Study on Artificial Intelligence (AI100) |url=https://ai100.stanford.edu/gathering-strength-gathering-storms-one-hundred-year-study-artificial-intelligence-ai100-2021-study |access-date=2025-06-02 |website=ai100.stanford.edu |language=en}} chaired by Michael Littman, was published in 2021.{{Cite web|last=Stacey|first=Kevin|date=16 Sep 2021|title=New Report Assesses Progress and Risks of Artificial Intelligence|url=https://hai.stanford.edu/news/new-report-assesses-progress-and-risks-artificial-intelligence|access-date=2022-01-20|website=Stanford University HAI News and Announcements|language=en}}{{Cite web|last=McKendrick|first=Joe|date=18 Sep 2021|title=Artificial intelligence success is tied to ability to augment, not just automate|url=https://www.zdnet.com/article/artificial-intelligence-success-is-tied-to-ability-to-augment-not-just-automate/|access-date=2022-01-20|website=ZDNet|language=en}}

He co-founded and has served as board chair of the Partnership on AI, a non-profit organization bringing together Apple, Amazon, Facebook, Google, DeepMind, IBM, and Microsoft with representatives from civil society, academia, and non-profit R&D. The organization's website points at initiatives, including studies of risk scores in criminal justice,{{Cite web|url=https://www.partnershiponai.org/report-on-machine-learning-in-risk-assessment-tools-in-the-u-s-criminal-justice-system/|title=Report on Algorithmic Risk Assessment Tools in the U.S. Criminal Justice System|date=2019-04-23|website=The Partnership on AI|language=en-US|access-date=2020-04-24}} facial recognition systems,{{Cite web|url=https://www.partnershiponai.org/facial-recognition-systems/|title=Bringing Facial Recognition Systems To Light|date=2020-02-18|website=The Partnership on AI|language=en-US|access-date=2020-04-24}} AI and economy,{{Cite web|url=https://www.partnershiponai.org/compendium-synthesis/|title=AI, Labor, and the Economy Case Study Compendium|date=2019-04-30|website=The Partnership on AI|language=en-US|access-date=2020-04-24}} AI safety,{{Cite web|url=https://www.partnershiponai.org/safelife/|title=Introducing SafeLife: Safety Benchmarks for Reinforcement Learning|date=2019-12-04|website=The Partnership on AI|language=en-US|access-date=2020-04-24}} AI and media integrity,{{Cite web|url=https://www.partnershiponai.org/ai-and-media-integrity-steering-committee/|title=AI and Media Integrity Steering Committee|date=2019-12-05|website=The Partnership on AI|language=en-US|access-date=2020-04-24}} and documentation of AI systems.{{Cite web|url=https://www.partnershiponai.org/about-ml/|title=About ML|website=The Partnership on AI|language=en-US|access-date=2020-04-24}}

He founded and chairs the Aether Committee at Microsoft, Microsoft's internal committee on the responsible development and fielding of AI technologies.{{Cite web|url=https://news.microsoft.com/2018/03/29/satya-nadella-email-to-employees-embracing-our-future-intelligent-cloud-and-intelligent-edge/|title=Satya Nadella email to employees: Embracing our future: Intelligent Cloud and Intelligent Edge|last=Nadella|first=Satya|date=2018-03-29|website=Microsoft Stories|access-date=3 June 2019}}{{Cite web|url=https://www.youtube.com/watch?v=tL7t2O5Iu8E&|title=Microsoft #TechTalk: AI and Ethics|date=2019-07-17|website=YouTube|access-date=2020-04-15}} He reported that the Aether Committee had made recommendations on and guided decisions that have influenced Microsoft's commercial AI efforts.{{Cite news|url=https://www.geekwire.com/2018/microsoft-cutting-off-sales-ai-ethics-top-researcher-eric-horvitz-says/|title=Microsoft is turning down some sales over AI ethics, top researcher Eric Horvitz says|last=Boyle|first=Alan|date=9 April 2018|work=GeekWire|access-date=3 June 2019}}{{Cite web|url=https://www.youtube.com/watch?v=PjJHPko4J2k|title=Conference on Ethics & AI: Keynote Session|date=9 April 2018|website=YouTube|publisher=Carnegie Mellon University|access-date=3 June 2019}} In April 2020, Microsoft published content on principles, guidelines, and tools developed by the Aether Committee and its working groups, including teams focused on AI reliability and safety, bias and fairness, intelligibility and explanation, and human-AI collaboration.{{Cite web|url=https://www.microsoft.com/en-us/ai/responsible-ai|title=Responsible AI principles from Microsoft|website=Microsoft Artificial Intelligence|access-date=2020-04-15}}

• {{Citation|last1=Horvitz|first1=E.|title=Computation and Action Under Bounded Resources|url=http://erichorvitz.com/horvitzdiss.pdf|type=Dissertation|location=Stanford, CA|publisher=Stanford University|date=December 1990}}

• {{Citation|last=Horvitz|first=E.|title=AI, people, and society|journal=Science|volume=357|issue=6346|pages=7|date=2017-07-07|doi=10.1126/science.aao2466|pmid=28684472|bibcode=2017Sci...357....7H|doi-access=free}}
• {{Citation|last1=Gershman|first1=S.|title=Computational rationality: A converging paradigm for intelligence in brains, minds, and machines|url=https://www.science.org/doi/10.1126/science.aac6076|volume=349|issue=6245|pages=273–278|doi=10.1126/science.aac6076|last2=Horvitz|first2=E.|last3=Tenenbaum|first3=J.|date=2015-07-17|journal=Science|pmid=26185246|bibcode=2015Sci...349..273G|s2cid=14818619|url-access=subscription}}
• {{Citation|last1=Kamar|first1=E.|isbn=978-0-9817381-1-6|last2=Hacker|first2=S.|last3=Horvitz|first3=E.|title=Combining Human and Machine Intelligence in Large-scale Crowdsourcing|url=http://research.microsoft.com/en-us/um/people/horvitz/CrowdSynth.pdf|journal=AAMAS '12 Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems - Volume 1|volume=1|pages=467–474|location=Richland, SC|publisher=International Foundation for Autonomous Agents and Multiagent Systems|date=June 2012}}
• {{Citation|last=Horvitz|first=E.|title=Artificial Intelligence in the Open World|url=http://research.microsoft.com/en-us/um/people/horvitz/AAAI_Presidential_Lecture_Eric_Horvitz.htm|journal=Opening Session of the Annual Meeting, Association for the Advancement of Artificial Intelligence|type=Lecture|location=Chicago, IL|date=July 2008}}
• {{Citation|last1=Horvitz|first1=E|title=Models of Attention in Computing and Communication: from Principles to Applications|date=March 2003|url=http://erichorvitz.com/Models_of_attention_in_computing.pdf|work=Communications of the ACM|volume=46|issue=3|pages=52–59|publisher=ACM|doi=10.1145/636772.636798|last2=Kadie|first2=C|last3=Paek|first3=T|last4=Hovel|first4=D|s2cid=2584780}}
• {{Citation|last=Horvitz|first=E.|title=Principles and Applications of Continual Computation|url=http://erichorvitz.com/ftp/cc_aij_horvitz.pdf|journal=Artificial Intelligence|volume=126|issue=1–2|pages=159–196|doi=10.1016/S0004-3702(00)00082-5|citeseerx=10.1.1.476.5653|date=February 2001}}
• {{Cite book|last=Horvitz|first=E|chapter=Principles of mixed-initiative user interfaces|date=May 1999|chapter-url=http://erichorvitz.com/chi99horvitz.pdf|pages=159–166|place=New York, NY|publisher=ACM|doi=10.1145/302979.303030|isbn=0-201-48559-1|title=Proceedings of the SIGCHI conference on Human factors in computing systems the CHI is the limit - CHI '99|s2cid=8943607}}
• {{Citation|last1=Horvitz|first1=E|title=Display of information for time-critical decision making|date=August 1995|url=http://erichorvitz.com/cognitive_cost_display_of_information_UAI_95.pdf|work=UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence|pages=296–305|place=San Francisco, CA|publisher=Morgan Kaufmann Publishers Inc|isbn=1-55860-385-9|last2=Barry|first2=M}}
• {{Citation|last1=D|first1=Heckerman|last2=Horvitz|first2=E|last3=Nathwani|first3=Bharat|title=Toward Normative Expert Systems: Part I, the Pathfinder Project|journal=Methods of Information in Medicine|url=http://erichorvitz.com/Toward_Normative_Systems_MIM.pdf|volume=31|pages=90–105|date=June 1992|issue=2|pmid=1635470|doi=10.1055/s-0038-1634867|s2cid=14672300 }}
• {{Citation|last3=Horvitz|first3=E.|last2=Breese|first2=J.|last1=Henrion|first1=M.|s2cid=18217939|title=Decision analysis and expert systems|journal=AI Magazine|volume=12|issue=4|pages=64–91|date=1991|location=Menlo Park, CA|publisher=American Association for Artificial Intelligence|doi=10.1609/aimag.v12i4.919}}
• {{Citation|last1=Shwe|first1=M.|last2=Middleton|first2=B.|last3=Heckerman|first3=D.|last4=Hernion|first4=M.|last5=Horvitz|first5=E.|last6=Lehmann|first6=H.|last7=Cooper|first7=G.|title=Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base|pages=241–255|url=http://erichorvitz.com/QMR-Decision-Theory-Evaluation.pdf|journal=Methods of Information in Medicine| volume=30 |issue=4| date=October 1991|doi=10.1055/s-0038-1634846|s2cid=2279911 }}
• {{Citation|last1=Horvitz|first1=E.|last2=Cooper|first2=G.F.|last3=Heckerman|first3=D.|title=Reflection and action under scarce resources: Theoretical principles and empirical study|url=http://erichorvitz.com/ij89.pdf|journal=IJCAI'89 Proceedings of the 11th International Joint Conference on Artificial Intelligence - Volume 2|pages=1121–1127|location=San Francisco, CA|publisher=Morgan Kaufmann Publishers Inc.| date=August 1989}}
• {{Citation|last=Horvitz|first=E.|title=Reasoning under varying and uncertain resource constraints|url=http://erichorvitz.com/ftp/AI88.PDF|journal=AAAI'88 Proceedings of the Seventh AAAI National Conference on Artificial Intelligence|pages=111–116|publisher=AAAI Press|date=August 1988}}
• {{Citation|last1=Horvitz|first1=E.|last2=Breese|first2=J.|last3=Henrion|first3=M.|title=Decision theory in expert systems and artificial intelligence|url=http://research.microsoft.com/en-us/um/people/horvitz/dt.pdf |journal=International Journal of Approximate Reasoning|volume=2|issue=3|pages=247–302|date=July 1988|doi=10.1016/0888-613X(88)90120-X|publisher=Elsevier Science Inc.|location=New York, NY}}
• {{Citation|last=Horvitz|first=E.|title=Reasoning about beliefs and actions under computational resource constraints|url=http://erichorvitz.com/U87.PDF|pages=429–447|location=Arlington, VA|publisher=AUAI Press|arxiv=1304.2759|date=July 1987|bibcode=2013arXiv1304.2759H|isbn=0-444-87417-8}}

• {{Citation|title=AI and Our Future With Machines with Dr. Eric Horvitz|url=https://www.microsoft.com/en-us/research/blog/ai-and-our-future-with-machines-eric-horvitz/|publisher=Microsoft Research Podcast|date=4 Dec 2017}}
• {{Citation|title=Potential and Pitfalls of AI with Dr. Eric Horvitz|url=https://www.microsoft.com/en-us/research/lab/microsoft-research-india/articles/podcast-potential-and-pitfalls-of-ai-with-dr-eric-horvitz/|publisher=Microsoft Research Podcast|date=5 Mar 2020}}

{{reflist|2}}

• [https://www.microsoft.com/en-us/research/people/horvitz/ Profile page at Microsoft Research]
• [https://ai100.stanford.edu/ One Hundred Year Study on Artificial Intelligence (AI100)]
• [http://research.microsoft.com/en-us/um/people/horvitz/Challange_Problems_for_AI.mp3 Audio: Challenge Problems for AI]
• [https://www.youtube.com/watch?v=dpoVh9xwdD4 TEDx Austin: Making Friends with Artificial Intelligence]
• [http://www.sciencefriday.com/segment/03/15/2013/improving-healthcare-one-search-at-a-time.html NPR: Science Friday: Improving Healthcare One Search at a Time] {{Webarchive|url=https://web.archive.org/web/20150924095347/http://www.sciencefriday.com/segment/03/15/2013/improving-healthcare-one-search-at-a-time.html |date=2015-09-24 }}
• [https://www.bbc.com/news/technology-26147990 BBC: "Artificial intelligence: How to turn Siri into Samantha"]
• [http://videolectures.net/kdd2014_horvitz_people_society/ Keynote address, Conference on Knowledge Discovery and Data Mining (SIGKDD), August 2014: Videolectures.net]

{{Authority control}}

__FORCETOC__

{{DEFAULTSORT:Horvitz, Eric}}

Category:Living people

Category:American computer scientists

Category:People in information technology

Category:Microsoft employees

Category:Fellows of the American Academy of Arts and Sciences

Category:Fellows of the Association for the Advancement of Artificial Intelligence

Category:Presidents of the Association for the Advancement of Artificial Intelligence

Category:Members of the American Philosophical Society

Category:Year of birth missing (living people)

Category:Microsoft technical fellows