Primary Sources and Historical Records

This is an annotated bibliography of primary documents in the history and governance of artificial intelligence: original papers, proposals, official reports and policy instruments, each linked to a verifiable record. It complements [LIT-FOUNDATIONAL-001], which covers technical literature from 1986 onward, by collecting earlier documents (1943-1976), long-run speculation, and public statements and legal instruments (2015-2025). Records were checked on 2026-10-15; annotations state only what the document's title, abstract or opening text supports.

Encyclopedic summaries often compress primary documents into one-line claims. Reading the originals matters because several widely repeated statements about AI history (what the Dartmouth proposal asked for, what the Lighthill report concluded, what ALPAC recommended) are easier to evaluate against the text. This index does not summarize those debates; it gives the reader the document. Interpretive accounts belong to [HIST-002], [HIST-003] and [HIST-004]. Inclusion is not endorsement.

Selection rule: a document is included if (a) it is cited in the encyclopedia's history, safety or governance articles or is standard in them, and (b) a bibliographic record or an official or archival copy could be reached and read at least in part on 2026-10-15. Several documents could not be reached and are therefore omitted; see section 7.

The documents fall into four periods.

1. Foundations of computation, information and neural models (1943-1950). McCulloch and Pitts (1943) on a logical calculus of nervous activity; Shannon (1948) on communication; Turing (1950) on machine intelligence.
2. Founding proposals and early programs (1955-1966). The Dartmouth proposal (McCarthy et al., 1955), Samuel (1959), Rosenblatt (1958), Weizenbaum (1966).
3. Critical and evaluative reports (1966-1976). The National Research Council's report on machine translation (1966), Minsky and Papert (1969), the Lighthill report (1972), Newell and Simon (1976). See [HIST-004] for the later interpretation of these as drivers of funding retrenchment; this article only lists the documents.
4. Long-run speculation and public instruments (1966, 1993, 2015-2025). Good (1966), Vinge (1993), and recent organizational statements, declarations and legal texts.

#4.1 Computation, information and neural models

• McCulloch & Pitts (1943). A logical calculus of the ideas immanent in nervous activity, Bulletin of Mathematical Biophysics. A mathematical model of networks of idealized neurons.
• Shannon (1948). A Mathematical Theory of Communication, Bell System Technical Journal. The foundation of information theory. See [FOUND-009] for the planned treatment of information theory.
• Turing (1950). Computing Machinery and Intelligence, Mind. Proposes replacing the question "Can machines think?" with an imitation game; see [FOUND-005].

#4.2 Founding proposals and early programs

• McCarthy, Minsky, Rochester & Shannon (1955). A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence, dated 31 August 1955. The opening text proposes "a 2 month, 10 man study of artificial intelligence" during the summer of 1956 at Dartmouth College, "on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it." See [HIST-002].
• Rosenblatt (1958). The perceptron: A probabilistic model for information storage and organization in the brain, Psychological Review. The original description of the perceptron model. See [HIST-003].
• Samuel (1959). Some Studies in Machine Learning Using the Game of Checkers, IBM Journal of Research and Development. An early machine-learning study using the game of checkers.
• Weizenbaum (1966). ELIZA - A Computer Program For the Study of Natural Language Communication Between Man and Machine, Communications of the ACM. Describes the ELIZA program.
• Newell & Simon (1976). Computer science as empirical inquiry, Communications of the ACM. A foundational statement of the symbolic approach; read it for the authors' own framing.

#4.3 Critical and evaluative reports

• National Research Council (1966). Language and Machines: Computers in Translation and Linguistics. National Academies Press, 138 pages. A report on machine translation and computational linguistics. Histories of AI funding cite it; readers should consult the report itself before relying on those summaries.
• Minsky & Papert (1969). Perceptrons. MIT Press. A mathematical analysis of perceptrons; an expanded edition appeared in 1988.
• Lighthill (1972). Artificial Intelligence: A General Survey, report to the Science Research Council, dated July 1972. Its introduction states that the Council had received "an increasing number of applications for research support" in the field and needed "a general view of the field". The hosted copy is a transcription by Chilton Computing, which also hosts commentary by Sutherland, Needham, Longuet-Higgins and Michie.

#4.4 Long-run speculation (Hypothetical)

• Good (1966). Speculations Concerning the First Ultraintelligent Machine, Advances in Computers. Often cited as an early statement of the intelligence-explosion idea. The idea is Hypothetical; see [ASI-EXP-001].
• Vinge (1993). The coming technological singularity: How to survive in the post-human era, NASA Lewis Research Center, Vision 21 symposium. An essay on a possible technological singularity (Hypothetical).

#4.5 Organizational statements, declarations and legal instruments

• OpenAI (2015). Introducing OpenAI, 11 December 2015. Founding announcement; see [ORG-LABS-001].
• Future of Life Institute (2017). Asilomar AI Principles. A set of principles that the Institute states were "developed at the Beneficial AI 2017 conference".
• National Institute of Standards and Technology (Tabassi, 2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0), released 26 January 2023. See [ORG-STANDARDS-001].
• Executive Order 14110 (2023). Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence, signed 30 October 2023, published at 88 FR 75191 on 1 November 2023. Revoked on 2025-01-20 according to the encyclopedia's editorial brief, which treats Executive Order 14148 (Initial Rescissions of Harmful Executive Orders and Actions, signed 20 January 2025, 90 FR 8237) as the rescinding order. [VERIFICATION NEEDED: the full text of EO 14148 was not read to confirm that it names EO 14110.] See [GOV-US-001].
• Bletchley Declaration (2023). The Bletchley Declaration by Countries Attending the AI Safety Summit, 1-2 November 2023, UK government. See [GOV-INTL-001].
• Center for AI Safety (n.d.). Statement on AI Risk: "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war." The page displays signatories; the statement's date was not displayed on the page read. See [ORG-SAFETY-001].
• European Parliament & Council (2024). Regulation (EU) 2024/1689 of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). For application dates and amendments see [GOV-EU-001]. [VERIFICATION NEEDED: whether the text has been amended since 2024; the version read was an archived copy]

(The total of 21 includes EO 14148, listed together with EO 14110 in section 4.5.)

How to use this index:

• Checking a claim in history. Before repeating a statement about the Dartmouth workshop, read the proposal's opening text (4.2). The proposal is a plan, not a record of what happened at the workshop; see [HIST-002].
• Checking a legal status. Policy instruments change. EO 14110's status is stated in 4.5; for the EU Act and U.S. policy, follow the links to [GOV-EU-001] and [GOV-US-001].
• Tracing an organization's stated purpose. Compare OpenAI (2015) with the later account in [ORG-LABS-001].

Strengths. Original texts, with identifiers, support independent reading.

Limitations.

• Access. Several documents expected in an index of this kind could not be reached on 2026-10-15 and are omitted: for example, the RAND publication page for H. Dreyfus's 1965 paper on AI returned an access block, and no other copy was checked. [VERIFICATION NEEDED: exact title and RAND report number] They are candidates for later versions, not claims about their existence.
• Copies. Some entries link to a host's copy (Chilton Computing for the Lighthill report; the Stanford site for the Dartmouth proposal) rather than the original publisher.
• Context. Annotations do not reproduce context or reception.
• Legal texts. Instruments may be amended or revoked after the review date.

Unresolved. Whether a conclusion commonly attributed to a report (for example, that the NRC and Lighthill reports caused funding reductions) is supported is a historical question addressed in [HIST-004], not here. It is Contested among historians as far as this article can say; no source on this is cited.

As of 2026-10-15 the 21 entries in section 4 have verified bibliographic records or readable official copies. Not every entry was read in full: the Dartmouth proposal and the Lighthill report were read at their opening text only, and the EU Act was verified at title level. The index has not been reviewed by a historian.

Living-record protocol: new entries are added when a document becomes standard in the encyclopedia; legal instruments get a dated status note at each review; documents known to be missing (section 7) are added when a reachable copy is found. Next review: 2027-04-15 for legal instruments (because they change), 2027-10-15 for the rest. No forecasts are made.

• Companion Literature:
  • LIT-FOUNDATIONAL-001 Foundational Papers Index: Technical literature from 1986.
• History:
  • HIST-001 Precursors to AI: Context for 4.1.
  • HIST-002 The Dartmouth Workshop: Context for the 1955 proposal.
  • HIST-003 Symbolic AI and the Perceptron: Context for 4.2 and 4.3.
  • HIST-004 AI Winters: Interpretation of the 1966-1973 reports.
• Governance:
  • GOV-LAW-001 AI Governance: Context for 4.5.
  • GOV-EU-001 The EU AI Act, GOV-US-001 United States AI Policy, GOV-INTL-001 International AI Governance and Compute Governance: Legal instruments in detail.
• Living References: TIME-001, FACT-001, LIB-001.