Downloads
AI OSI Downloads
Free reference material, templates, and examples for documenting AI-influenced decisions.
Readable by non-experts. Defensible under scrutiny.
No sign-ups. No endorsements.
Decision & Governance Templates
Core, human-readable records for documenting AI-influenced decisions before memory fades or context is lost.
AI Decision Record
A minimal, time-bounded record that a decision occurred, who owned it, what AI influenced it, and whether it is still valid.
→ Download Google Doc (5 pages · 5 KB)
AI Change Record
A record of material changes to AI systems or their use, and why continued operation was allowed.
→ Download Google Doc (5 pages · 4 KB)
AI Legitimacy Check
A short pre-decision check to determine whether an AI-enabled activity should proceed at all.
→ Download Google Doc (5 pages · 5 KB)
AI Incident Record
A first-capture record of AI-related issues, complaints, or unexpected outcomes, before hindsight sets in.
→ Download Google Doc (4 pages · 4 KB)
AI System Record
A standing record of what AI systems exist, what they are used for, and who is accountable for them.
→ Download Google Doc (5 pages · 4 KB)
Global Example Decision Records
Plain-language examples showing how AI decision records apply to everyday situations that affect real people.
Healthcare Decision Record
A record of a clinical or care-related decision affecting a person’s health, including who decided, what information was relied on, and when the decision must be revisited.
→ Download Google Doc (5 pages · 5 KB)
Employment Decision Record
A record of a decision affecting hiring, promotion, discipline, or termination, documenting criteria used, human approval, and any automated influence.
→ Download Google Doc (5 pages · 5 KB)
Financial Access Decision Record
A record of a decision affecting credit, insurance, eligibility, pricing, or financial access, making assumptions, uncertainty, and review paths explicit.
→ Download Google Doc (5 pages · 6 KB)
Registers & Dashboards
Simple indexes and summaries for seeing patterns, drift, and accumulated risk across decisions and systems.
AI Decision & System Register
An index of AI systems and AI-influenced decisions, linking records, owners, status, and review dates.
→ Download Google Sheet (1 page · 2 KB)
AI Change & Incident Register
A chronological log of AI system changes, scope shifts, and incidents over time.
→ Download Google Sheet (1 page · 2 KB)
AI Governance Overview & Signals
An executive-level dashboard showing trends, distributions, and items requiring attention.
→ Download Google Sheet (2 pages · 3 KB)
Canonical Specifications
Reference architectures defining how AI OSI works, how records relate, and how evidence holds up over time.
AI OSI Core v0.1: Decision Evidence & Temporal Validity Protocol
A universal, descriptive standard for recording and expiring AI-influenced decisions.
→ Download PDF (5 pages · 118.8 KB)
AI OSI Stack: Canonical Specification (v4)
The authoritative layered governance architecture defining evidence expectations for high-stakes AI systems.
→ Download PDF (72 pages · 392.6 KB)
Alignment & Transparency
Non-binding materials showing how AI OSI artifacts map to external expectations and public accountability.
EU AI Act Annex IV Alignment Note
An illustrative, non-binding mapping between AI OSI artifacts and Annex IV documentation expectations.
→ Download PDF (4 pages · 80 KB)
Roadmap & Work Completed
A dated, versioned record of publications, milestones, and external evaluation activity.
→ Download PDF (14 pages · 52 KB)
Role Briefings
Short briefings translating AI OSI concepts for specific oversight and accountability roles.
Defensible AI Oversight for Boards & Audit Committees
A briefing on fiduciary exposure, oversight failure modes, and contemporaneous evidence.
→ Download PDF (3 pages · 75 KB)
Audit-Grade Evidence for Regulators & Oversight Bodies
A briefing on consistent, reviewable governance evidence across institutions and systems.
→ Download PDF (3 pages · 75 KB)
AI Governance Evidence for Legal & Regulatory Defense
A briefing on discovery-ready evidence and temporal legitimacy under audit, litigation, or enforcement.
→ Download PDF (3 pages · 76 KB)
Non-Operational AI Governance for CISOs & Cyber Risk Leaders
A briefing on establishing AI governance evidence without disrupting security operations.
→ Download PDF (3 pages · 76 KB)
Notices
These materials are descriptive, not prescriptive. They do not provide legal advice, certification, regulatory guidance, or a declaration of compliance. Use remains the responsibility of the adopting institution and its accountable officers.
Canonical specifications are archived on Zenodo with persistent DOIs for independent timestamping and version history.