Massachusetts Institute of Technology

The Massachusetts Institute of Technology (MIT) is a private research university in Cambridge, Massachusetts. It was incorporated in 1861, opened to students in Boston in 1865, and moved to its present Cambridge campus in 1916. MIT is organized around five schools and the MIT Stephen A. Schwarzman College of Computing. Its work in computing, artificial intelligence, robotics, and related fields is distributed across academic departments, laboratories, and Institute-wide initiatives rather than housed in one organization.^{[1][2][16][17][20]}

MIT has been a major institutional center of AI research since the late 1950s, but it was not the site where the term "artificial intelligence" was introduced. The 1955 proposal that used the phrase was written for a summer project at Dartmouth College. At the time, John McCarthy was affiliated with Dartmouth, Marvin Minsky with Harvard University, Nathaniel Rochester with IBM, and Claude Shannon with Bell Telephone Laboratories. McCarthy and Minsky later established coordinated AI research at MIT in 1959. That effort developed into the MIT Artificial Intelligence Laboratory and, after a 2003 merger with the Laboratory for Computer Science, the Computer Science and Artificial Intelligence Laboratory (CSAIL).^[8][9][10]

MIT's historical contributions include the Whirlwind real-time computer, the Lisp programming language, early natural-language programs such as ELIZA and SHRDLU, work in computer vision, and behavior-based robotics. Its present-day AI activity spans CSAIL, the Department of Electrical Engineering and Computer Science (EECS), the Schwarzman College, the MIT Media Lab, MIT Lincoln Laboratory, the Siegel Family Quest for Intelligence, the MIT Jameel Clinic, and other groups. These entities have different purposes and governance. They should not be treated as interchangeable names for MIT or as a single consolidated AI laboratory.^{[6][9][16][20]}

Institution

Origins and campus

Natural scientist William Barton Rogers advocated a polytechnic institution that would connect scientific principles with practical work. The Commonwealth of Massachusetts chartered MIT on April 10, 1861. The American Civil War delayed the start of instruction, and the first students attended classes in rented Boston space in 1865. Laboratory teaching and "learning by doing" were central to the educational model. MIT admitted its first woman student in 1871 and remained an independent, coeducational institution.^[1][2]

MIT initially occupied buildings in Boston's Back Bay. It crossed the Charles River to Cambridge in 1916, where a group of interconnected academic buildings became the core of the campus. The Institute reports a 168-acre Cambridge campus. The physical campus includes teaching and research facilities, residences, athletic areas, and public art, while major MIT operations also exist elsewhere. MIT Lincoln Laboratory, for example, is a federally funded research and development center headquartered in Lexington, Massachusetts, and is not a department on the Cambridge campus.^[2][33]

MIT's development into a large research university accelerated during and after the Second World War. Its engineering history includes wartime radar research, postwar electronics and computing, and the growth of federally sponsored laboratories. This history matters to AI because early interactive computing at MIT arose from a broader institutional network of electrical engineering, control, information theory, military research, and time-sharing rather than from an isolated AI program.^[6][10][12]

Governance and scale

MIT is governed by the MIT Corporation, its board of trustees. The Corporation approves major policies and budgets, elects the president, and exercises fiduciary oversight. Academic authority is also shared through the faculty and the Institute's schools, departments, and committees. Sally Kornbluth became MIT's eighteenth president in January 2023.^[5][35]

For the 2025-2026 academic year, MIT reported 11,816 students: 4,561 undergraduates and 7,255 graduate students. It also reported 3,437 international students in degree programs. In October 2025, MIT counted 1,087 faculty members. Its employee total, including Lincoln Laboratory, was 17,033.^[3][4]

Selected institutional figures available at the research cutoff are:

Scroll sideways for more →

The distinction between endowment value, annual revenue, and research expense is important. MIT's fiscal 2025 operating statement listed $5.373 billion in revenue and $5.117 billion in expenses. Sponsored research was the largest expense category at $2.206 billion, but MIT does not identify all computing or AI activity as a single line item. It would therefore be misleading to infer an "AI budget" from the Institute's overall research spending.^[4]

These figures also have different reporting dates. Enrollment is an academic-year snapshot, faculty and employee counts use specified census dates, and financial measures cover a fiscal year. They cannot be combined into a single real-time profile without preserving those dates. Rankings and award totals are omitted here because they use changing methodologies and do not explain how MIT's AI organizations relate to one another. The institutional counts provide context for scale, not a measure of research quality or priority.

Computing and AI history

Whirlwind and real-time computing

MIT's path toward interactive computing predates its formal AI program. The Whirlwind project began during the Second World War as an attempt to build a flight-training simulator. The design evolved into an electronic digital computer built for real-time operation. Whirlwind became operational in the early 1950s, and an improved magnetic-core memory was installed in 1953. The system contributed to interactive display and control techniques and became an important technical ancestor of the SAGE air-defense network.^[6][7]

Whirlwind should be described with care. Claims that it was simply "the first computer" obscure competing definitions and projects. Its historically important features were its emphasis on reliable real-time response, interactive input and display, and the practical development of magnetic-core memory. The SAGE work also helped motivate the establishment of MIT Lincoln Laboratory in 1951. Lincoln Laboratory is operated by MIT as a federally funded research and development center, but its national-security mission, sponsor relationships, personnel systems, and Lexington site distinguish it from an academic department.^[6][7][33]

Dartmouth proposal and the MIT AI Project

The proposal for the Dartmouth Summer Research Project on Artificial Intelligence is dated August 31, 1955. It proposed a two-month study at Dartmouth during the summer of 1956 and advanced the conjecture that features of learning and intelligence could be described precisely enough for machines to simulate them. The proposal is a primary source for the phrase "artificial intelligence" and for the project's intended research program.^[8]

The affiliations printed on the proposal establish a boundary often blurred in accounts centered on MIT. McCarthy was then at Dartmouth, Minsky at Harvard, Rochester at IBM, and Shannon at Bell Telephone Laboratories. The gathering took place at Dartmouth. McCarthy and Minsky subsequently joined MIT, but their later positions do not make the Dartmouth project an MIT event. MIT's importance lies in the sustained research organization they helped build afterward.^[8][9]

In 1959, McCarthy and Minsky started the first coordinated AI research at MIT. The group worked through the Research Laboratory for Electronics and the Computation Center. It used the IBM 704 and pursued symbolic computation, problem solving, language, perception, and machine learning. The effort is often called the MIT Artificial Intelligence Project or AI Group. It was initially a research group, not yet the independent laboratory that later carried the AI Lab name.^[9]

McCarthy developed Lisp during this period. His 1960 paper described a programming system for symbolic expressions that had been developed for the IBM 704 by MIT's AI group. Lisp's list structures, recursive functions, symbolic evaluation, and ability to represent programs as data made it influential in AI research and programming-language design. The paper supports the institutional connection without requiring the broader claim that every early Lisp implementation or later dialect originated at MIT.^[11]

Project MAC, the AI Laboratory, and LCS

Project MAC began at MIT in 1963 with support from the Advanced Research Projects Agency, later known as DARPA. It combined work on time-sharing, interactive computing, programming systems, and machine intelligence. The National Academies' history of federal computing research credits Project MAC with an important role in time-sharing while cautioning that it did not create the first time-sharing system. Its participants worked on the Compatible Time-Sharing System and later Multics, alongside the AI group and other projects.^[10]

Project MAC's name was interpreted in several ways, including Multi-Access Computer and Machine-Aided Cognition. The umbrella organization brought together research programs that were technically and administratively related but not identical. AI work received substantial support through Project MAC, while the laboratory also pursued operating systems, computer-aided design, networks, and broader computer science.^[10][12]

Administrative boundaries changed over time. MIT's own progress report for July 1970 through July 1971 states that the Artificial Intelligence group became an independent MIT laboratory in December 1970. Project MAC continued, and the Laboratory for Computer Science (LCS) grew from the non-AI side of that lineage. Accounts that give 1969, 1970, or 1971 for "the AI Lab" may be referring to different stages: organizational separation, formal independence, or a reporting period. December 1970 is the clearest date in the contemporary progress report for independent-laboratory status.^[9][12]

The AI Laboratory and LCS remained separate for more than three decades. Their work increasingly overlapped as AI came to depend on systems, theory, networks, security, and large-scale software, while computer science incorporated machine learning and intelligent interfaces. MIT merged the two laboratories on July 1, 2003 to form CSAIL.^[9]

Selected contributions and their limits

MIT laboratories produced or hosted many important computing projects, but an institutional history should not turn every affiliation into sole credit. The following examples have strong primary documentation and illustrate different periods:

Scroll sideways for more →

ELIZA is sometimes called an early chatbot. Weizenbaum's paper described a method for conducting natural-language conversation through decomposition rules, reassembly rules, and a script. The program's best-known script imitated a nondirective psychotherapist, which made simple transformations appear responsive. Its historical significance lies partly in how readily users attributed understanding to a system with narrow mechanisms. Presenting ELIZA as a direct technical ancestor of contemporary large language models without explaining these differences would overstate continuity.^[13]

SHRDLU connected parsing, reference resolution, planning, and action in a simulated world of blocks. Its restricted vocabulary and environment were essential to its performance. The project showed how multiple components could support a coherent interaction when knowledge and possible actions were tightly bounded. It did not establish a general solution to natural-language understanding, and its limitations helped clarify how difficult open-world language and common-sense reasoning would be.^[14]

Brooks's behavior-based robotics work challenged the assumption that a robot required a complete internal world model before it could act. His layered architecture allowed lower-level behaviors to continue operating while higher layers modified or suppressed outputs. The paper described both a control architecture and a particular experimental program. Its influence on later robotics does not mean that planning, representation, or learning became unnecessary; modern systems often combine reactive control with estimation, mapping, optimization, and learned components.^[15]

MIT research also contributed to vision, knowledge representation, programming languages, parallel and distributed systems, and human-computer interaction. Those histories involve numerous laboratories and institutions. Broad statements that MIT "invented computer vision," "founded neural networks," or single-handedly created academic AI are not supported by the evidence and erase work at Dartmouth, Carnegie Mellon, Stanford, IBM, Bell Labs, and other centers.^[8][9][10]

Current computing and AI organization

MIT's present computing organization is deliberately distributed. The Schwarzman College provides an Institute-level academic structure; EECS is a department; CSAIL and the Media Lab are research laboratories; Lincoln Laboratory is a federally funded research and development center; and the Quest, Jameel Clinic, and industry partnerships are initiatives with narrower missions. A person or project may participate in more than one of these structures.^[16][17][20]

Schwarzman College and EECS

MIT announced the Stephen A. Schwarzman College of Computing in October 2018. The plan described a $1 billion commitment enabled by a $350 million foundational gift from Stephen Schwarzman. It created a sixth major academic unit alongside MIT's five schools and called for 50 additional faculty positions, split between core computing and positions jointly based in other fields. The college was intended to strengthen computer science and AI while connecting computing with disciplines across MIT and examining social and ethical implications.^[16]

The college did not absorb every computing laboratory or replace the schools. EECS is jointly part of the School of Engineering and the Schwarzman College. It is organized into faculty areas that include electrical engineering, computer science, and Artificial Intelligence and Decision-Making (AI+D). The undergraduate program in Artificial Intelligence and Decision Making, Course 6-4, combines computing, optimization, probability, inference, and decision-making coursework.^[17][18]

The college's Building 45 on Vassar Street opened in early 2024. The building brought together college leadership and groups from several computing units. A shared building does not make those units administratively identical; it is infrastructure for collaboration and teaching as well as organizational space.^[19]

Research laboratories and initiatives

CSAIL describes itself as MIT's largest research laboratory. Its published figures list more than 1,700 members, over 130 principal researchers, about 1,200 graduate students, 119 postdoctoral researchers, 60 research groups, and more than 900 active projects. These are the laboratory's own changing counts rather than a census of all AI researchers at MIT. CSAIL's research spans AI, systems, theory, human-computer interaction, robotics, security, and other areas. Daniela Rus has served as its director since 2012.^[20]

The MIT Media Lab opened in 1985 in the School of Architecture and Planning. It grew around research that crossed media, design, computation, sensing, interfaces, and human experience. Historically associated work includes wearable computing, tangible interfaces, and affective computing. The Media Lab is not a division of CSAIL, although faculty, students, and projects may collaborate across the two.^[21]

The MIT Quest for Intelligence launched in 2018 as an Institute-wide effort to study intelligence and apply resulting tools. It was renamed the Siegel Family Quest for Intelligence in 2025 following a major gift from the Siegel Family Endowment. Official Quest material identifies 2025, not 2024, as the renaming year and does not support attaching an unverified gift amount to the name change.^[22]

MIT and IBM created the MIT-IBM Watson AI Lab in 2017 under a ten-year, $240 million IBM commitment. In April 2026, the partners announced the MIT-IBM Computing Research Lab as a new ten-year collaboration spanning AI, algorithms, and quantum computing. The new lab was described as an evolution of the Watson AI Lab. At the July 28, 2026 research cutoff, it is therefore inaccurate to list the Watson lab alone as if the 2017 arrangement remained MIT's current partnership name without noting the transition.^[23][24]

The MIT Jameel Clinic was founded in 2018 through a partnership with Community Jameel to support machine learning in health. It coordinates work across computer science, engineering, medicine, and the life sciences but is an Institute initiative rather than a hospital or medical school. Projects include diagnostics, clinical decision tools, and molecular discovery.^[25]

MIT Lincoln Laboratory applies AI and machine learning within national-security research programs, including sensing, communications, cyber operations, aerospace, and decision support. Its work is part of MIT's research enterprise, but project disclosure can be constrained by sponsor and security requirements. Public annual reports provide examples rather than a complete inventory.^[33][34]

MIT launched the Generative AI Impact Consortium in February 2025 to connect faculty across its schools and college with industry partners. Initial seed funding supported research on areas such as health, manufacturing, climate, education, and human-AI interaction. The consortium is a funding and collaboration mechanism, not a new department or a single foundation-model laboratory.^[29]

The main institutional boundaries can be summarized as follows:

Scroll sideways for more →

Research themes and examples

AI, systems, and robotics

MIT's contemporary AI research covers neural networks, language and speech, vision, robotics, optimization, probabilistic modeling, algorithms, and the interaction between AI systems and people. CSAIL also works on operating systems, networks, programming languages, cryptography, theory, and security. Treating all CSAIL work as AI understates the laboratory's computer-science scope; treating all MIT AI work as CSAIL misses research in other departments and laboratories.^[17][20]

Robotics illustrates this overlap. A robot may require perception, state estimation, planning, control, mechanical design, embedded systems, learning, and human interaction. MIT groups working in these areas sit in CSAIL, Mechanical Engineering, Aeronautics and Astronautics, the Media Lab, Lincoln Laboratory, and other units. Historical work from Brooks's lab helped establish behavior-based approaches, while current work includes soft and modular robots, autonomous vehicles, manipulation, assistive systems, and multi-robot coordination. Project claims should be attributed to the responsible group rather than to "MIT AI" as a single actor.^[15][20]

The same caution applies to industrial outcomes. CSAIL's official spin-off list includes companies such as Akamai, Boston Dynamics, iRobot, and RSA Security. It is evidence of connections to laboratory research and personnel, not proof that CSAIL alone created every product made by those companies. Boston Dynamics and iRobot, for example, developed through separate companies with their own teams, financing, partners, and later owners.^[30]

Machine learning in health and molecular discovery

Work associated with the Jameel Clinic has provided visible examples of machine learning used in molecular screening. In a 2020 Cell paper, researchers trained a directed-message-passing neural network on compound structures and antibacterial activity. They used it to screen the Drug Repurposing Hub and identified halicin as an antibacterial candidate structurally different from conventional antibiotics. Laboratory tests found broad activity, and the compound showed efficacy in mouse infection models.^[26]

Halicin was not thereby established as an approved medicine. The study reported preclinical discovery and testing, not human clinical safety or effectiveness. Its significance was methodological: a learned molecular representation helped prioritize a compound that conventional similarity-based searches might not have selected. The result also depended on experimental validation; the model's prediction was a screening step, not a substitute for microbiology or clinical trials.^[26]

A 2023 Nature Chemical Biology paper applied a similar workflow to Acinetobacter baumannii, a pathogen for which new treatments are needed. The researchers screened roughly 7,500 molecules, identified abaucin, investigated a mechanism involving lipoprotein trafficking, and reported activity in a mouse wound model. Abaucin was a narrow-spectrum experimental lead, not a general antibiotic or approved therapy. The work shows both the promise and boundary of AI-assisted discovery: computational selection narrowed a search, while biological experiments established the reported activity.^[27]

These studies should not be summarized as "AI invented antibiotics." The compounds, models, assays, and mechanistic studies came from interdisciplinary teams. The published evidence concerns specified datasets, organisms, and experimental conditions. Translation from a research result to a safe, manufacturable, clinically effective drug remains a separate process.^[26][27]

Measurement, fairness, and social effects

MIT research has also examined how automated systems perform across demographic groups. The 2018 Gender Shades study by Joy Buolamwini and Timnit Gebru evaluated three commercial gender-classification systems using a dataset balanced by skin type and gender. It reported error rates as high as 34.7 percent for darker-skinned women, while the maximum error rate for lighter-skinned men was 0.8 percent across the three tested systems.^[28]

The study did not evaluate every face-recognition task or prove that all later versions retained the same error rates. It tested particular commercial classifiers, categories, and images at a particular time. Its lasting contribution was to demonstrate an intersectional evaluation method and to show that aggregate accuracy could conceal large subgroup differences. It also illustrates why institutional AI work includes measurement, governance, and public consequences as well as model capability.^[28]

The Schwarzman College's founding plan explicitly included social and ethical dimensions of computing. Such work also appears in the humanities and social sciences, management, architecture and planning, and cross-Institute programs. No one center represents MIT's full position on AI policy. Individual faculty can disagree, and research findings should not automatically be presented as official Institute policy.^[16]

Generative AI and industry collaboration

The Generative AI Impact Consortium was designed to support cross-Institute projects rather than concentrate funding only in computer science. In June 2025, MIT reported that 55 proposals had been selected for the consortium's inaugural seed grants, with additional projects selected for company funding.^[29]

Industry relationships supply funding, data, engineering expertise, and routes to deployment, but they also require clear attribution. MIT reported $176 million in direct industry sponsorship in fiscal 2025 and relationships with more than 400 organizations. The same MIT source described industry support as about 20 percent of research expenditures. Because $176 million is not 20 percent of the Institute-wide $2.206 billion sponsored-research expense, that percentage evidently uses a narrower reporting boundary and should not be compared directly with the Institute-wide figure. These figures cover all fields rather than AI alone.^[4][30]

The MIT-IBM collaborations are one example of a named partnership. The 2017 Watson AI Lab was funded by IBM and brought MIT and IBM researchers into joint projects. The 2026 Computing Research Lab extended the relationship into another ten-year phase and added a stated emphasis on quantum computing. Outputs from such partnerships should be identified as jointly sponsored or partnership-reported rather than treated as independent measures of MIT's research superiority.^[23][24]

Education and training

MIT students do not enroll in an independent "AI school." Undergraduate and graduate education is organized through departments and interdisciplinary programs. EECS offers computing programs, including the Bachelor of Science in Artificial Intelligence and Decision Making, Course 6-4. The program combines computer science and mathematical foundations with methods for machine learning, inference, optimization, and decisions. It is one route through MIT's curriculum, not a count of everyone who studies AI.^[17][18]

Related study also occurs through other departments and interdisciplinary programs. Graduate students receive degrees from academic programs even when much of their research takes place in a laboratory such as CSAIL.^[17][20][21]

Course numbers and offerings change. The current Course 6-4 catalog, for example, lists 6.3900 as Introduction to Machine Learning. Readers should therefore use the current catalog rather than treat a historical course-number list as permanent.^[18]

Laboratories also train students through research assistantships, theses, internships, seminars, and project teams. CSAIL's membership figures include a large graduate-student population, while the Media Lab and other units admit or host students through their academic homes. Laboratory membership and degree enrollment are different measurements; adding them together would double-count people with multiple affiliations.^[20][21]

Translation, funding, and accountability

Technology transfer and companies

MIT's Technology Licensing Office manages invention disclosure, patents, licensing, and related technology-transfer processes. In fiscal 2025, MIT reported 684 invention disclosures, 623 United States patent applications, 282 issued United States patents, and support for 38 companies based on MIT intellectual property. These Institute-wide figures span fields from biotechnology to materials and software. They should not be represented as an AI-only output count.^[30]

Company formation is only one form of research translation. Open-source software, standards, clinical collaborations, government systems, trained graduates, and published methods can matter without creating a spin-off. Conversely, a company founded by an MIT alumnus is not necessarily an MIT spin-off. A defensible connection requires evidence such as licensed intellectual property, a documented laboratory origin, or a company and university account of the relationship.^[30]

Selected CSAIL-connected companies include Akamai, Boston Dynamics, iRobot, and RSA Security. The laboratory's spin-off page is a useful institutional record but is promotional rather than a complete economic-impact study. It should support bounded examples, not unsourced totals for employment, valuation, venture funding, or revenue.^[30]

Sponsored research and institutional responsibility

MIT receives substantial support from the federal government, industry, foundations, and donors. Funding source can shape a project's questions, disclosure rules, and application environment, while academic publication and conflict-of-interest rules create additional constraints. Lincoln Laboratory's national-security sponsorship differs from an open academic grant, and an industry consortium differs from a philanthropic gift. Collapsing these arrangements into a generic "MIT-funded" label hides important governance boundaries.^[4][30][33]

The Institute's handling of donations from Jeffrey Epstein is a documented case of governance failure relevant to the Media Lab and MIT more broadly. An independent fact-finding report released in 2020 found that Epstein made 10 donations to MIT totaling $850,000 between 2002 and 2017. Nine donations totaling $750,000 came after his 2008 conviction, including $525,000 for the Media Lab. The report found significant errors in judgment and failures of policy and leadership, although it did not find that senior leaders violated then-existing written policies. MIT subsequently selected four nonprofits serving survivors of sexual abuse and exploitation to receive gifts totaling $850,000, and later reported that committees were refining recommendations on outside engagements and gift acceptance.^[31][32]

This episode should neither be omitted from an institutional account nor allowed to consume the history of unrelated research. It demonstrates that laboratories operate within Institute governance and that funding decisions can create institutional responsibility. It also illustrates why a page about MIT must distinguish evidence from official self-description and include independent review when assessing contested conduct.^[31][32]

MIT Technology Review presents another boundary. The publication says it was founded at MIT in 1899 and that its coverage is editorially independent despite MIT ownership. Its journalism should therefore not be cited as an official MIT institutional position merely because "MIT" appears in its name.^[36]

Scope and related subjects

This article is about the institution whose canonical AI Wiki slug is mit and whose visible title is "Massachusetts Institute of Technology." Several related names refer to distinct subjects:

• The MIT License is a permissive software license. It has its own article, MIT License, and should not be treated as an organizational unit, a research program, or a statement of MIT's current licensing policy.
• The MIT Media Lab and CSAIL are separate laboratories inside MIT. Their projects and histories may be discussed here when they illuminate the institution, but they are not synonyms for MIT or for each other.^[9][21]
• The Schwarzman College is an academic college; EECS is a department shared by that college and the School of Engineering; AI+D is a faculty area within EECS.^[16][17]
• The Siegel Family Quest for Intelligence, MIT Jameel Clinic, Generative AI Impact Consortium, and MIT-IBM Computing Research Lab are initiatives or partnerships with bounded missions. None is the sole successor to all earlier MIT AI work.^{[22][24][25][29]}
• MIT Lincoln Laboratory is operated by MIT but is a federally funded research and development center with a national-security mission and a Lexington headquarters.^[33][34]
• MIT Technology Review is MIT-owned but editorially independent. Its articles are publications about technology, not automatically Institute policy.^[36]
• The separate MIT "GenAI Divide" report (2025) article concerns a particular report and controversy, not the university as a whole.

The distinction also applies to people and companies. A faculty appointment, degree, postdoctoral position, licensed invention, and corporate partnership are different relationships. The page should state which relationship is documented instead of using "MIT-affiliated" to imply institutional authorship or endorsement.

At the research cutoff of July 28, 2026, MIT remained a large, decentralized research university with a historically important AI lineage. Its record includes foundational work, current research across many units, education and technology transfer, sponsored partnerships, and governance failures. Describing that record accurately requires both recognition of MIT's influence and restraint about priority, exclusivity, and institutional boundaries.

See also

• Artificial intelligence
• History of artificial intelligence
• John McCarthy
• Marvin Minsky
• Lisp
• Computer vision
• Machine learning
• Robotics
• MIT License
• MIT "GenAI Divide" report (2025)

References

1. ^Massachusetts Institute of Technology. "Origins." facts.mit.edu/origins
2. ^Massachusetts Institute of Technology. "About MIT." mit.edu/about
3. ^Massachusetts Institute of Technology. "Enrollment statistics." MIT Facts. facts.mit.edu/enrollment-statistics
4. ^Massachusetts Institute of Technology. "MIT Facts: Employees, faculty, and operating financials." facts.mit.edu/employees ; facts.mit.edu/...faculty ; facts.mit.edu/operating-financials
5. ^MIT Corporation. "About the Corporation." corporation.mit.edu/about-corporation
6. ^MIT School of Engineering. "History." engineering.mit.edu/...history
7. ^MIT Lincoln Laboratory. "SAGE: Semi-Automatic Ground Environment air defense system." ll.mit.edu/...round-environment-air-defense-system
8. ^McCarthy, John, Marvin Minsky, Nathaniel Rochester, and Claude Shannon. "A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence." August 31, 1955. www-formal.stanford.edu/...dartmouth
9. ^MIT Computer Science and Artificial Intelligence Laboratory. "Mission and history"; Massachusetts Institute of Technology. "MIT's computer science, AI labs merge." csail.mit.edu/...mission-history ; news.mit.edu/...mits-computer-science-ai-labs-merge
10. ^National Research Council. Funding a Revolution: Government Support for Computing Research. National Academies Press, 1999. nationalacademies.org/...chapter
11. ^McCarthy, John. "Recursive Functions of Symbolic Expressions and Their Computation by Machine, Part I." Communications of the ACM 3, no. 4 (1960): 184-195. www-formal.stanford.edu/...recursive
12. ^Massachusetts Institute of Technology. Project MAC Progress Report VIII, July 1970-July 1971. bitsavers.org/...gress_Report_08_197007-197107.pdf
13. ^Weizenbaum, Joseph. "ELIZA - A Computer Program for the Study of Natural Language Communication Between Man and Machine." Communications of the ACM 9, no. 1 (1966): 36-45. doi.org/...365153.365168
14. ^Winograd, Terry. Procedures as a Representation for Data in a Computer Program for Understanding Natural Language. MIT AI Technical Report 235, 1971. eric.ed.gov
15. ^Brooks, Rodney A. "A Robust Layered Control System for a Mobile Robot." IEEE Journal of Robotics and Automation 2, no. 1 (1986): 14-23. doi.org/...JRA.1986.1087032
16. ^Massachusetts Institute of Technology. "MIT reshapes itself to shape the future." MIT News, October 15, 2018. news.mit.edu/...hwarzman-college-of-computing-1015
17. ^MIT Department of Electrical Engineering and Computer Science. "About EECS" and "Departmental organization." eecs.mit.edu/about
18. ^Massachusetts Institute of Technology. "Artificial Intelligence and Decision Making (Course 6-4)." MIT Catalog. catalog.mit.edu/...ence-decision-making-course-6-4
19. ^Massachusetts Institute of Technology. "A crossroads for computing at MIT." MIT News, April 11, 2024. news.mit.edu/...crossroads-computing-mit-0411
20. ^MIT Computer Science and Artificial Intelligence Laboratory. "CSAIL by the numbers" and "Leadership." cap.csail.mit.edu/...csail-numbers ; csail.mit.edu/...leadership
21. ^MIT Media Lab. "History." media.mit.edu/...history
22. ^MIT Siegel Family Quest for Intelligence. "Our roots in the Quest." sqi.mit.edu/...roots-quest
23. ^Massachusetts Institute of Technology. "IBM and MIT to pursue joint research in artificial intelligence, establish new MIT-IBM Watson AI Lab." MIT News, September 7, 2017. news.mit.edu/...n-artificial-intelligence-lab-0907
24. ^IBM. "The MIT-IBM Computing Research Lab launches to shape the future of AI and quantum computing." April 29, 2026. newsroom.ibm.com/...re-of-ai-and-quantum-computing
25. ^MIT Jameel Clinic. "About MIT Jameel Clinic." jclinic.mit.edu/about-mit-jameel-clinic
26. ^Stokes, Jonathan M., et al. "A Deep Learning Approach to Antibiotic Discovery." Cell 180, no. 4 (2020): 688-702. doi.org/...j.cell.2020.01.021
27. ^Liu, Gary, et al. "Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii." Nature Chemical Biology 19 (2023): 1342-1350. doi.org/...s41589-023-01349-8
28. ^Buolamwini, Joy, and Timnit Gebru. "Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification." Proceedings of Machine Learning Research 81 (2018): 77-91. proceedings.mlr.press/...buolamwini18a
29. ^Massachusetts Institute of Technology. "Introducing the MIT Generative AI Impact Consortium" and "Researchers present bold ideas for AI at MIT Generative AI Impact Consortium event." MIT News, February 3 and June 20, 2025. news.mit.edu/...nerative-ai-impact-consortium-0203 ; news.mit.edu/...ve-ai-impact-consortium-event-0620
30. ^Massachusetts Institute of Technology. "MIT and industry"; MIT Computer Science and Artificial Intelligence Laboratory. "Spin-offs." facts.mit.edu/mit-industry ; csail.mit.edu/...spin-offs
31. ^Massachusetts Institute of Technology. "FAQ on the fact-finding report on MIT and Jeffrey Epstein"; "MIT announces recipients of Epstein-related funds"; and "Letter to the MIT community: Fulfilling MIT's commitments." MIT News, January 10 and 27 and December 3, 2020. news.mit.edu/...inding-report-jeffrey-epstein-0110 ; news.mit.edu/...rofits-sexual-abuse-survivors-0127 ; news.mit.edu/...letter-commitments-community-1203
32. ^Goodwin Procter LLP. Report concerning Jeffrey E. Epstein's interactions with the Massachusetts Institute of Technology. January 2020. facultygovernance.mit.edu/...dwinProcterReport.pdf
33. ^MIT Lincoln Laboratory. "Organization." ll.mit.edu/...organization
34. ^MIT Lincoln Laboratory. 2024 Annual Report. June 2025. ll.mit.edu/...MITLL_2024_Annual_Report.pdf
35. ^Massachusetts Institute of Technology. "About Sally Kornbluth." Office of the President. president.mit.edu/about
36. ^MIT Technology Review. "Who we are." CyberSecure event site, 2021. event.technologyreview.com/...mittr