AI Responsibility Statement
Our commitment. This course was developed with the assistance of artificial intelligence tools in the drafting, structuring, and editing of course content. We are committed to transparency about how AI was used, what human oversight was applied, and what this means for the accuracy and reliability of what you are reading. Every factual claim in this course has been subject to human review and, where relevant, cross-referenced against primary sources, regulatory publications, and verifiable enforcement records. We believe that a course on surveillance and market integrity has a particular obligation to model the standards of accuracy and transparency that it teaches. This statement is part of that obligation.
1. How AI was used in developing this course
The course content was developed using large language model (LLM) AI tools, specifically Anthropic’s Claude, in combination with the judgement and subject-matter expertise of the working trade surveillance practitioners who built the course. AI was used in four capacities: as a research aggregation tool, compiling and organising publicly available information from regulatory publications, enforcement orders, academic literature, and industry sources; as a drafting assistant, producing initial versions of explanatory prose that were then reviewed, edited, and in many cases substantially rewritten; as an editorial assistant, identifying inconsistencies, suggesting improvements to structure and clarity, and flagging areas requiring factual verification; and as a formatting tool, producing document templates and structured layouts.
AI was not used as a substitute for expert judgement on legal, regulatory, or technical matters. All characterisations of law, regulation, and enforcement outcomes were reviewed against primary sources. Where AI-generated drafts contained unsupported assertions, those assertions were either verified and sourced, or removed. Where AI-generated content expressed legal or regulatory positions with unwarranted confidence, that content was revised to reflect appropriate epistemic caution.
Narration, video, and imagery. The narration in the course videos is synthetic speech. Each script is written and reviewed by the course team against the module text before recording, and the recording is then produced by a text-to-speech system running on the course team’s own equipment. No human voice is recorded and no real person’s voice is imitated. The motion graphics in the videos are built from text and data approved by the course team. A small number of module introductions carry decorative background illustrations produced with an image-generation tool; they depict no real person, place, organisation, or event, and they carry no data.
2. The limits of AI-generated content and how we addressed them
Large language models are capable of generating plausible-sounding but factually incorrect statements, a phenomenon commonly described as “hallucination”. In a course on trade surveillance, where the accuracy of enforcement case details, statutory citations, and regulatory standards directly affects professional practice, hallucinated content would not merely be an academic failing; it could cause real harm to practitioners who rely on it.
To address this risk, the following standards were applied throughout development. Every enforcement case cited in this course is drawn from a primary source: a published regulatory order, court judgement, or official regulator publication, cited with enough specificity for the reader to locate the source document independently. No enforcement case is described solely on the basis of AI-generated summaries. Every statutory citation names the relevant act, section, and jurisdiction and has been verified against the text of the legislation; where legislation has been amended, the applicable version is reflected. Figures cited in the course, whether penalty amounts, market statistics, or dates, carry source attributions, and learners are encouraged to verify figures independently where data may have been updated after the course’s stated review dates.
The course content also went through a multi-stage editorial process beyond ordinary drafting review: structural and editorial review, cross-reference and terminology consistency checks, regulatory accuracy verification against primary sources, adversarial audits of factual claims, and repeated correction rounds, each documented internally. Despite these controls, no content production process is infallible. We correct substantiated errors promptly in the live course content.
3. Scope and limits: where learners should exercise caution
Visualisations, assessment questions, and illustrative examples are designed to support understanding of concepts and do not constitute legal, regulatory, or investment advice. Scenario-based examples are illustrative and may simplify the facts of real cases for educational purposes; learners undertaking professional investigation or enforcement work should consult primary source documents rather than relying on course summaries.
The regulatory landscape covered in this course evolves continuously. Legislation is amended, enforcement priorities shift, and new guidance is issued on an ongoing basis. Content reflects the regulatory position as at the verification dates stated in the course’s regulatory currency notices; learners operating in regulated environments are responsible for ensuring they are aware of subsequent developments.
This course does not provide legal advice, and nothing in it should be construed as such. Learners facing specific legal or regulatory questions should consult qualified legal counsel in the relevant jurisdiction.
4. Human oversight and editorial accountability
This course was built, reviewed, and is maintained by working trade surveillance practitioners. All module content was reviewed by the course team before publication, and characterisations of law, regulation, and enforcement outcomes were checked against primary sources as described above. Editorial accountability for the course rests with Irontide. For readers in the European Union, this statement is also the disclosure contemplated by Article 50(4) of Regulation (EU) 2024/1689 (the EU Artificial Intelligence Act): the AI-assisted text in this course has undergone human review and fact-checking against primary sources, and Irontide Academy holds editorial responsibility for it. Enquiries about that responsibility may be sent to hello@irontide.academy.
The use of AI tools in this course was deliberate and bounded. The decision to disclose that use fully reflects a commitment to the same principles of transparency that underpin market integrity: participants in any system, whether a financial market or an educational programme, are entitled to know the nature of the information they are relying on and how it was produced.
5. Reporting errors and providing feedback
We actively solicit error reports from learners. If you identify a factual inaccuracy, a misstatement of law or regulation, an outdated reference, or any other content concern, please report it to support@irontide.academy with the module and lesson reference. Reports are reviewed by the course team and, where substantiated, corrections are applied to the live course content.
Feedback that improves the accuracy of the course benefits every subsequent learner. We treat such contributions as a form of professional collaboration.
Version 2.0, effective from first public release. This statement will be updated if the course’s AI-assisted development practices change materially, or if the tools used change in ways that affect the standards described above. Questions about this statement: hello@irontide.academy.