Carnegie Mellon University

RawGraph

Carnegie Mellon University (CMU) is a private research university in Pittsburgh, Pennsylvania.[1][6] Andrew Carnegie funded the Carnegie Technical Schools in 1900; the first class entered in 1906, the institution began granting bachelor's degrees as Carnegie Institute of Technology in 1912, and the 1967 merger with the Mellon Institute created Carnegie Mellon University.[1] CMU is especially prominent in the history of computer science, artificial intelligence, speech recognition, machine learning, and robotics.[3]

CMU's place in early AI is tied to a collaboration, not to a program written wholly at the university. In 1956, Allen Newell and J. C. Shaw of the RAND Corporation and Herbert A. Simon of Carnegie Institute of Technology reported the Logic Theory Machine, commonly called the Logic Theorist. Their paper states that the experiments ran on RAND's JOHNNIAC computer.[2] CMU later established a Computer Science Department in 1965, launched the Robotics Institute in 1979, elevated computer science to a school in 1988, and created the first academic department devoted to machine learning in 2006.[3][4][5]

Figures in this article are dated because enrollment, finances, organizational structure, and research programs change. As of September 15, 2025, CMU reported 16,582 enrolled students: 7,728 undergraduates and 8,854 graduate students.[6] Its audited endowment totaled approximately $3.485 billion at June 30, 2025.[7] In March 2026, the university reported more than $727 million in fiscal-year 2025 research expenditures, including just under $456 million from federal sources and $133 million of institutional support.[8]

Origins and institutional development

Andrew Carnegie donated $1 million to Pittsburgh in 1900 for a technical school serving working-class men and women. The Carnegie Technical Schools initially offered two- and three-year programs in the arts and engineering-related fields and included a college for women. The first class was admitted in 1906.[1] In 1912, the institution became Carnegie Institute of Technology and began awarding bachelor's degrees. It merged with the Mellon Institute, a science research center founded by the Mellon family, in 1967.[1]

Computing at Carnegie Tech developed before computer science became a separate department. An IBM 650 arrived in 1956, and Herbert Simon and Alan Perlis established a computation center. The university's historical account dates an interdisciplinary Systems and Communications Sciences doctoral program to 1961. In 1965, a $5 million grant from the R. K. Mellon Foundation supported a Computer Science Department headed by Perlis; at first it admitted doctoral students rather than undergraduates.[3]

The present School of Computer Science (SCS) emerged from that department. CMU's Faculty Senate approved college status in 1988, A. Nico Habermann became the first dean, and the school began operations in December 1988; the university issued its public announcement in January 1989.[3] These dates explain why CMU sources variously describe SCS as established in 1988 or announced in 1989.

Early artificial intelligence

Logic Theorist and symbolic problem solving

The Logic Theorist was designed to find proofs in propositional logic. The surviving October 1956 report identifies Newell and Shaw with RAND and Simon with Carnegie Tech, and says its empirical trials were conducted on JOHNNIAC at RAND.[2] CMU's own historical timeline describes Newell, Simon, and Shaw as the three developers and calls the system one of the first AI programs.[9] This evidence supports describing the project as an important CMU-RAND collaboration; it does not support the narrower claim that the program was written or run only at CMU.

Newell and Simon later received the 1975 ACM A. M. Turing Award. The ACM citation credits their joint work in artificial intelligence, the psychology of human cognition, and list processing.[10] CMU's institutional histories treat their work as the beginning of a local tradition connecting symbolic computation with empirical theories of human problem solving.[3][9]

Building computer science as an institution

CMU's historical importance also reflects a sequence of organizational investments. The Computer Science Department began in 1965; the Robotics Institute followed in 1979; and SCS brought the Computer Science Department, Robotics Institute, Center for Machine Translation, and researchers from the Information Technology Center together at college level in 1988.[3] The school later added units for human-computer interaction, language technologies, machine learning, computational biology, and software and societal systems.[3][11]

As of July 2026, SCS listed seven departments:[11]

  • Computer Science Department
  • Human-Computer Interaction Institute
  • Language Technologies Institute
  • Machine Learning Department
  • Ray and Stephanie Lane Computational Biology Department
  • Robotics Institute
  • Software and Societal Systems Department

The Software Engineering Institute (SEI) is not an eighth SCS department. It is a federally funded research and development center sponsored by the U.S. Department of Defense and operated by CMU. The Defense Department established the SEI in 1984, operations began in early 1985, and the operating contract was renewed in June 2025 for 2025-2030 with a five-year extension option.[12]

Research traditions

The examples below are selected, sourceable milestones rather than a claim that CMU originated every field in which its researchers worked.

Speech and language technology

CMU speech research in the 1970s included the Hearsay, Harpy, and Dragon systems. A 1989 research summary by Raj Reddy described Sphinx, demonstrated in 1987, as a speaker-independent system for continuous, large-vocabulary speech recognition; the same source records a real-time implementation in 1988 and the use of dialog and semantic knowledge in the Minds system.[13] These are historically bounded descriptions of the systems reported at the time, not comparisons with modern speech models.

The Center for Machine Translation was one of the units incorporated into SCS in 1988 and later became the Language Technologies Institute.[3] The institute describes its origin as machine-translation research and its subsequent expansion to speech, natural-language processing, information retrieval, and multimodal forms of communication.[14] In a 1997 paper, the JANUS III team reported a multilingual speech-to-speech translation system and published performance for both transcribed and recognized speech under the study's test conditions.[15]

Robotics and autonomous vehicles

CMU's Robotics Institute was launched in October 1979 with a five-year, $5 million commitment from Westinghouse.[4] SCS identifies Raj Reddy as founding director and describes the institute as the first academic department dedicated exclusively to robotics.[16] That institutional "first" concerns a dedicated academic unit; robotics research and teaching existed elsewhere before 1979.

The Navlab program illustrates both the progress and limits of 1990s autonomous driving. During the 1995 "No Hands Across America" trip, the RALPH road-following system steered Navlab 5 for 2,797 of 2,849 miles, or 98.2% of the route.[17] Rahul Sukthankar's 1997 CMU dissertation explains the boundary of that statistic: the system maintained lane position, while the human driver changed lanes and speed for short-term goals. The percentage therefore measured distance under automated steering, not complete autonomous driving.[18]

In the 2007 DARPA Urban Challenge, six of 11 finalists completed the staged urban course. CMU-led Tartan Racing placed first under DARPA's scoring rules and received the $2 million prize.[19] DARPA's retrospective explicitly cautioned that the competitions' effect on later commercial and defense deployment is difficult to quantify, so the win should not be treated as proof of a simple causal path from one university project to an industry.[19]

Machine learning and game solving

The Machine Learning Department was founded in spring 2006 after evolving from the Center for Automated Learning and Discovery, created in 1997. The department describes itself as the world's first academic machine-learning department and records earlier specialized degree programs through the center.[5] This is an organizational first, not a claim that machine-learning research began at CMU.

Two projects illustrate different strands of later research:

  • The Never-Ending Language Learner (NELL) was presented as a case study in continual or "never-ending" learning. Its 2015 AAAI paper reported that, after running since January 2010, the system had accumulated more than 80 million confidence-weighted beliefs together with learned features and parameters.[20] The number describes the state reported in that paper; it is not an accuracy score or a count of independently verified facts.
  • Libratus applied computational game theory to heads-up no-limit Texas hold'em. The peer-reviewed 2018 report states that it defeated four specialist professionals over a 120,000-hand competition and describes a blueprint strategy, subgame solving, and a self-improvement component.[21] The result was for that two-player poker setting and does not demonstrate general intelligence.

Education

SCS began a dedicated undergraduate computer-science program during the 1989-1990 academic year after earlier students followed an applied mathematics and computer-science route.[3] The 2018 AI-degree announcement described instruction drawing on SCS and faculty from CMU's humanities and social sciences, public policy, and engineering colleges.[22]

In May 2018, CMU announced a Bachelor of Science in Artificial Intelligence beginning that fall and described it as the first undergraduate AI degree offered by a U.S. university. The announced curriculum combined computer science and mathematics with statistics, machine learning, symbolic computation, ethics, and social responsibility.[22] CMU reported that the first two students in the SCS program graduated in May 2020; its initial cohort contained 18 students.[23] "First" here refers to the university's stated category of a dedicated undergraduate AI degree, not to the first AI course, concentration, or individually designed degree.

CMU's own histories disagree on the founding year of the Human-Computer Interaction Institute. The SCS history lists the unit under 1993, while HCII's dated twentieth-anniversary account says it was founded in 1994.[3][24] The anniversary account describes its interdisciplinary work across design, computer science, and behavioral and social sciences.[24] The sources establish the institute's interdisciplinary scope but do not resolve the one-year chronology discrepancy.

Research infrastructure and initiatives

As of the article's July 28, 2026 cutoff, several recent programs extended CMU's work into physical AI, computer-assisted mathematics, and AI-supported education:

InitiativeVerifiable status by the cutoffEvidence boundary
Robotics Innovation CenterCMU officially opened the 150,000-square-foot facility at Hazelwood Green on February 27, 2026.[25]The opening and facility size are established; future scientific or economic impact is not.
Fujitsu-Carnegie Mellon Physical AI Research CenterCMU and Fujitsu announced the center in April 2026, with research spanning robotics, machine learning, language technologies, human-computer interaction, engineering, and philosophy.[26]The source is a joint institutional announcement and describes planned research, not completed outcomes.
Institute for Computer-Aided Reasoning in Mathematics (ICARM)CMU announced the NSF- and Simons Foundation-supported, three-year pilot in August 2025, with researchers in mathematics, philosophy, computer science, and partner institutions.[27]Its mission concerns formal methods, machine learning, and mathematical reasoning; projected applications are goals rather than demonstrated results.
LearnviaCMU announced the nonprofit learning collaborative in January 2026 and reported that faculty and students at 38 U.S. institutions were using its AI-enhanced Calculus I courseware during that academic year.[28]The announcement establishes launch and reported participation, not causal evidence of improved completion or achievement.

These initiatives show how CMU combines research groups across organizational boundaries. They should be evaluated through later peer-reviewed results, independent program assessments, and clearly dated operating data rather than through launch announcements alone.

Interpreting CMU's AI record

Several recurring qualifications are important when using CMU sources:

  • Collaborative attribution: Logic Theorist involved Newell and Shaw at RAND and Simon at Carnegie Tech, and its documented experiments ran at RAND.[2]
  • Definitions of "first": CMU sources describe firsts in specific institutional categories, such as a dedicated robotics department, a machine-learning academic department, and a dedicated U.S. undergraduate AI degree.[5][16][22] They do not imply the first research, course, or degree touching those subjects under every possible definition.
  • Experimental scope: Navlab's 98.2% figure refers to automated steering distance, while Libratus's result concerns a defined two-player poker competition.[18][21]
  • Institutional self-reporting: Enrollment and audited financial statements are suitable for dated descriptive figures, while announcements of centers and collaborations establish their existence and intended scope but not their eventual effectiveness.[6][7][25][26][27][28]

This distinction between documented history, measured experimental results, and institutional claims is necessary for evaluating CMU's influence without turning a university article into promotional material.

See also

References

  1. ^Carnegie Mellon University. "History." cmu.edu/...history
  2. ^Allen Newell, J. C. Shaw, and H. A. Simon. "Empirical Explorations of the Logic Theory Machine." RAND paper P-951-A, October 10, 1956. iiif.library.cmu.edu/...00004_fld00294_doc0001.pdf
  3. ^Carnegie Mellon University School of Computer Science. "A History of SCS." cs.cmu.edu/...history
  4. ^Carnegie Mellon University Robotics Institute. "RI History." ri.cmu.edu/...ri-history
  5. ^Carnegie Mellon University Machine Learning Department. "About Machine Learning at CMU: History of the Machine Learning Department." ml.cmu.edu/about
  6. ^Carnegie Mellon University. "Common Data Set 2025-2026," section B1, September 15, 2025. cmu.edu/...cds_2025_full_pdf.pdf
  7. ^Carnegie Mellon University. "Annual Financial Report 2025," note 7. cmu.edu/...annual-report-2025.pdf
  8. ^Farnam Jahanian. "2026 State of the University Remarks: As Prepared," Carnegie Mellon University, March 23, 2026. cmu.edu/...2026SOTURemarks.pdf
  9. ^Carnegie Mellon University. "125 Years of Carnegie Mellon University." cmu.edu/...cmu-history
  10. ^Association for Computing Machinery. "Allen Newell: A. M. Turing Award Laureate." amturing.acm.org/...newell_3167755.cfm
  11. ^Carnegie Mellon University School of Computer Science. "Departments." cs.cmu.edu/...departments
  12. ^Software Engineering Institute. "Software Engineering Institute Marks 40 Years of Innovation and a Renewed Contract with Defense Department." June 24, 2025. sei.cmu.edu/...ed-contract-with-defense-department
  13. ^Raj Reddy. "Speech Research at Carnegie Mellon." Proceedings of the Workshop on Speech and Natural Language, 1989, pp. 119-120. aclanthology.org/H89-1014
  14. ^Carnegie Mellon University Language Technologies Institute. "About LTI." lti.cmu.edu/about-lti
  15. ^Alon Lavie, Alex Waibel, Lori Levin, Donna Gates, Torsten Zeppenfeld, and Puming Zhan. "JANUS III: Speech-to-speech Translation in Multiple Languages." ICASSP 1997, vol. 1, pp. 99-102. publications.ri.cmu.edu/...n-in-multiple-languages
  16. ^Carnegie Mellon University School of Computer Science. "About SCS." cs.cmu.edu/about
  17. ^Todd M. Jochem. "Vision Based Tactical Driving." CMU-RI-TR-96-14, Carnegie Mellon University Robotics Institute, 1996. publications.ri.cmu.edu/...jochem_todd_1996_1.pdf
  18. ^Rahul Sukthankar. "Situation Awareness for Tactical Driving." CMU-RI-TR-97-08, Carnegie Mellon University Robotics Institute, 1997. publications.ri.cmu.edu/...hankar_rahul_1997_1.pdf
  19. ^Defense Advanced Research Projects Agency. "The DARPA Grand Challenge: Ten Years Later." March 13, 2014. darpa.mil/...grand-challenge-ten-years-later
  20. ^Tom M. Mitchell et al. "Never-Ending Learning." Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015. doi.org/...aaai.v29i1.9498
  21. ^Noam Brown and Tuomas Sandholm. "Superhuman AI for heads-up no-limit poker: Libratus beats top professionals." Science 359, no. 6374 (2018): 418-424. doi.org/...science.aao1733
  22. ^Carnegie Mellon University. "Carnegie Mellon Launches Undergraduate Degree in Artificial Intelligence." May 10, 2018. cmu.edu/...ai-undergraduate-degee
  23. ^Carnegie Mellon University. "CMU's AI Undergraduate Program Will Confer Its First Degrees." May 13, 2020. cmu.edu/...first-ai-degrees
  24. ^Carnegie Mellon University Human-Computer Interaction Institute. "Here's to Another 20 Years!" November 26, 2014. hcii.cmu.edu/...heres-another-20-years
  25. ^Carnegie Mellon University. "A New Home for Robotics and AI Discovery." February 27, 2026. cmu.edu/...opening
  26. ^Carnegie Mellon University School of Computer Science and Fujitsu. "Fujitsu, CMU Launch Joint Center for Physical AI." April 23, 2026. hcii.cmu.edu/...fujitsu-cmu-joint-center-physical-ai
  27. ^Carnegie Mellon University. "New NSF Institute at CMU Will Help Mathematicians Harness AI and Advance Discoveries." August 4, 2025. cmu.edu/...ians-harness-ai-and-advance-discoveries
  28. ^Carnegie Mellon University. "Introducing Learnvia." January 29, 2026. cmu.edu/...introducing-learnvia

Improve this article

Add missing citations, update stale details, or suggest a clearer explanation. Every suggestion is reviewed for sourcing before it goes live.

4 revisions · v5 · 2,452 words · full history

Fact-checks are independent of edits: a reviewer re-verifies the article against its sources and stamps the date. How we verify

Research and drafting on this wiki are AI-assisted, under named human editorial standards. How AI is used here

Reviewer note: Independent 2026-07-28 fact-check: 28 primary institutional or professional sources and 61 citation calls checked; root review rechecked Logic Theorist attribution, SCS and HCII chronology, Robotics Institute funding, Navlab steering scope, the seven-department and SEI distinction, dated enrollment and financial figures, four 2025-2026 initiatives, all six retained internal targets, and desktop/mobile production rendering.

Cite this page: AI Wiki. "Carnegie Mellon University." aiwiki.ai, updated 30 Jul 2026, fact-checked 30 Jul 2026. CC BY 4.0. https://aiwiki.ai/wiki/cmu

Suggest edit