Ten-school policy record and evidence correction

UChicago tool classifications corrected: directory approval is not institutional recommendation; Chrome AI extensions require specific-use review. Existing evidence, access conditions and assessment rules are preserved. This is a Tracker correction, not a new University policy.

Historical correction: counts and claims reflect this version, not the latest selected record.

2026-10-15T02:32:16.611Z · v6-chicago-tools-20261014

• California Institute of Technology (Caltech) · 31 reviewed current records · View policy page
  • Added records: 0; Corrected records: 0; Historical / held: 0

Previous V6 release history

Record revision, scope and original evidence

Original evidence and audience, course and unit scopes are preserved. Tracker correction time is not a policy effective date.

ac16401e0d7bc2dba928daede61333e76566d8760c4007fb34645ba4a977c2ba

{
  "addedClaimIds": [],
  "updatedClaimIds": [],
  "heldClaimIds": []
}

• cl-california-institute-of-technology-claude-56 — Anthropic Claude Premium is listed for California Institute of Technology (Caltech) in an official university AI tools source. Derived availability: conditionally allowed. Derived endorsement type: institutionally licensed or procured.
• cl-california-institute-of-technology-claude-57 — Anthropic Claude Standard is listed for California Institute of Technology (Caltech) in an official university AI tools source. Derived availability: conditionally allowed. Derived endorsement type: institutionally licensed or procured.
• cl-california-institute-of-technology-microsoft-copilot-58 — Microsoft Copilot Web is listed for California Institute of Technology (Caltech) in an official university AI tools source. Derived availability: allowed. Derived endorsement type: officially endorsed.
• cl-california-institute-of-technology-microsoft-copilot-for-m365-59 — Caltech IMSS lists its Windows 11 Copilot service as available and free. The official table labels that entry “Microsoft 365 Copilot for Windows 11”; this is kept as the source’s original label and does not establish availability of the separate Microsoft 365 Copilot for Education service.
• cl-california-institute-of-technology-zoom-ai-companion-60 — Zoom AI Companion is listed for California Institute of Technology (Caltech) in an official university AI tools source. Derived availability: allowed. Derived endorsement type: officially endorsed.
• cc-caltech-admissions-ethics-2 — Caltech admissions prohibits applicants from copying and pasting directly from an AI generator, relying on AI-generated content to outline or draft essays, replacing their unique voice with AI-generated content, or translating essays via AI.
• cc-caltech-admissions-ethics-3 — Caltech admissions permits applicants to use AI tools like Grammarly or Microsoft Editor for grammar and spelling review of completed essays, to generate brainstorming questions or exercises, and to research the college application process.
• cc-caltech-hss-genai-1 — In Caltech's Humanities and Social Sciences (HSS) division, students may use generative AI tools only in ways explicitly allowed by the course instructor in the course materials. Any usage not specifically allowed should be assumed to be disallowed.
• cc-caltech-hss-genai-2 — The Caltech HSS generative AI policy applies to all assignments including major papers, exams, discussion board posts, reflections, and problem sets.
• cc-caltech-hss-genai-3 — Caltech HSS students are expected to follow specific course guidance for documenting any permitted use of generative AI tools.
• cc-caltech-admissions-ethics-5 — Caltech states that academic AI-use policies for enrolled students are determined by individual departments and/or faculty. Institutional IMSS security and data-governance standards also apply to Caltech-data workloads; departmental autonomy does not establish that no university-wide rules exist.
• cc-caltech-hss-genai-4 — The Caltech HSS generative AI policy states that generative AI may promote learning in some contexts and impede it in others, and instructors prioritize student learning in setting these policies.
• draft-rule-imss-prohibition-consumer-ai-restricted-data — IMSS guidance prohibits restricted or confidential Caltech information in consumer AI tools. The LLM Standards permit a specific IMSS exception with compensating controls; catalogue access alone does not establish sensitive-data approval.
• draft-rule-imss-not-system-of-record — Generative AI platforms are strictly not approved for long-term storage, official records management, or institutional data retention; all records requiring retention must be stored in approved Caltech systems.
• draft-rule-imss-four-guiding-principles — All generative AI usage at Caltech should adhere to four institutional principles: prompt disclosure or referencing of GenAI tools and application plug-ins as applicable, data and information protection, content responsibility by the user, and the Caltech Honor Code. This guidance does not mandate a universal prompt-log format or tool-version record.
• draft-rule-imss-llm-standard-bedrock-sagemaker — For Caltech-data workloads, primary LLM inference must use Amazon Bedrock or other IMSS-approved platforms with no provider training on Caltech prompts/data, and custom training/fine-tuning must be performed on Amazon SageMaker or IMSS-approved secured environments.
• draft-rule-imss-llm-standard-bedrock-sagemaker-exception — Public, consumer, or unsanctioned GenAI tools are prohibited for Restricted or Confidential Caltech data unless IMSS issues a specific exception with compensating controls.
• draft-rule-imss-llm-standard-private-networking-kms — All AI traffic with Caltech data must use private networking (for example VPC Endpoints/PrivateLink or Direct Connect), with encryption in transit and at rest (TLS 1.2+; KMS CMKs). The examples are not an exclusive network list. Exceptions require Research Compliance Officer, CIO and CISO (or delegate) approval, compensating controls and an agreed sunset date.
• draft-rule-imss-llm-standard-agent-controls — AI agents operating in Caltech environments must be restricted to approved tools with explicit scopes and least-privilege credentials, and production tools must not allow arbitrary code execution without sandboxing.
• draft-rule-imss-llm-standard-human-subjects-irb — LLM usage involving human subjects or protected research data must receive prior Institutional Review Board (IRB) and IMSS data-use review.
• draft-rule-imss-llm-standard-vendor-dpa-no-training — External LLM vendors (non-AWS) must execute a Security & Data Processing Addendum that explicitly prohibits provider model training on Caltech data.
• draft-rule-imss-llm-standard-enforcement — Violations may result in access revocation, removal of workloads and disciplinary actions per Caltech IMSS. Persistent violations are escalated to the CIO, CISO, OGC and appropriate governance bodies.
• draft-rule-honorcode-ero-process — Undergraduate respondents offered the Early Resolution Option have 7 calendar days to decide whether to accept responsibility and waive a full Board of Control hearing, and may seek the support of a trained community advisor.
• draft-rule-honorcode-disability-accommodation — Students with disabilities in handbook processes may request reasonable accommodation from CASS. Complaint details need not be disclosed except insofar as they assist determination. Arrangements are individual and must be requested in a timely manner.
• draft-rule-honorcode-advisor-rights — Respondents in conduct proceedings have the right to an advisor from the Caltech community (who may not be an attorney); Student Affairs staff, Board of Control representatives, and Conduct Review Committee members are trained to serve as advisors.
• draft-rule-teaching-ctlo-detection-advisory — Caltech Undergraduate and Graduate Deans, CTLO, and the Hixon Writing Center strongly discourage the use of AI-detection tools for student writing due to false positives, false negatives, bias, and unreliability.
• draft-rule-bi1-coursework-policy — For Caltech Bi 1 (2026), generative AI tools (such as ChatGPT, Copilot, and Colab AI) are prohibited unless an assignment explicitly instructs students to engage with them on a per-problem basis.
• draft-rule-admissions-trusted-adult-benchmark — Caltech admissions advises applicants that a good rule of thumb for generative AI is that if a trusted adult shouldn't do it, AI shouldn't do it either.
• draft-rule-honorcode-graduate-and-research-misconduct — Under the Caltech Honor System, the Dean of Graduate Studies reviews graduate coursework violations (consulting the Graduate Honor Council), while allegations of research misconduct are handled separately by the Office of Research Compliance (ORC).
• draft-rule-imss-copilot-m365-education-coming-soon — Microsoft 365 Copilot for Education is listed in the official Caltech IMSS AI tools catalog as Coming Soon at $216/user/year via PTA order in access.caltech; it is not yet available as a live verified campus service.
• clm-caltech-admissions-fall-2027-ai-obligations — For the Fall 2027 application cycle, Caltech requires all applicants to review its ethical-use-of-AI guidelines before submitting supplemental essays. Noncompliance with those guidelines may result in rescission of admission. This is an admissions requirement for applicants, not a coursework rule or an automatic sanction for enrolled students.

Specific institute-wide policy governing generative AI in graduate doctoral dissertations and master's theses was not retrieved in this audit / not verified in public sources.

Dedicated institute-wide policy on instructor use of AI for grading student work was not retrieved in this audit / not verified in public sources.

Conflict between Dean written notification date and receipt of notification letter as the definitive trigger for the 7-calendar-day appeal clock.