# AI Literacy: Ada, AI and Verification

> An open, self-study course on checking what generative AI produces: compare
> the output with something established independently of it, find the first
> place they differ, name the failure, and claim no more than the check
> covered. For first-year undergraduates in any discipline; no programming
> required.

The verification skill below is a set of instructions a user can give an AI
system, so that it verifies its own work the way this course teaches and
leaves the user able to verify it. It does not make the system's output
correct.

## Skill

- [Verification skill](https://ai.qyber.dev/edu-ai-literacy-ada-verification/course/verification-skill.md): instructions for an AI system, as Markdown with skill frontmatter, for users to install as a skill or supply as custom instructions or a system prompt
- [Using the skill](https://ai.qyber.dev/edu-ai-literacy-ada-verification/course/ai-skills.html): the user's half -- how to install it and how to check what comes back

## Course

- [The course in brief](https://ai.qyber.dev/edu-ai-literacy-ada-verification/course/content-note.md): the content of every unit in compact form, as Markdown
- [Home](https://ai.qyber.dev/edu-ai-literacy-ada-verification/index.html): routes through the course, units and material
- [Unit 0: Introduction and how to use this course](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-00-introduction.html): How to use the course, what AI systems produced, and the learning contract you are agreeing to.
- [Unit 1: What generative AI is for](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-01-what-generative-ai-is-for.html): Five classes of task sorted by where the reference lives, and why that -- not the topic -- decides what checking costs.
- [Unit 2: Ada's ordered procedure](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-02-ada-ordered-procedure.html): Lovelace, the Analytical Engine, Note G, and what it takes to specify a computation exactly.
- [Unit 3: What a language model does differently](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-03-language-models.html): Next-token prediction set against deterministic execution, and why plausibility is not correctness.
- [Unit 4: Choosing the tool](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-04-choosing-the-tool.html): Four kinds of system, what each one changes about the checking you owe, and what a query costs in money, time and energy.
- [Unit 5: Prompts, context and attachments](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-05-prompts-context-attachments.html): What actually reaches the system, why the window has an edge you cannot see, and why an attachment makes an answer checkable rather than true.
- [Unit 6: Meet the Bernoulli numbers](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-06-bernoulli-numbers.html): Enough mathematics to know what correct output looks like — including the two convention traps that make correct programs disagree.
- [Unit 7: Ask AI to generate the routine](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-07-generate-routine.html): A deliberately underspecified prompt, and the plausible mistakes it produces.
- [Unit 8: Verification lab](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-08-verification-lab.html): Run generated code against an exact oracle, find the first divergence, and name the failure mode.
- [Unit 9: Grounding and citation checking](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-09-grounding-citation.html): The course's second worked case, run on a question with no oracle -- what a citation establishes, and six named ways one fails.
- [Unit 10: Re-prompting, specification, and overreliance](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-10-reprompting.html): Turning a vague request into testable requirements. Most clauses are bugs that already happened.
- [Unit 11: Agents and automated actions](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-11-agents-automated-actions.html): What changes when a system acts rather than answers -- an action can change the world before you have looked at it, so sort by what recovery costs and check the calls rather than the report.
- [Unit 12: High-stakes use and escalation](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-12-high-stakes-escalation.html): A second axis -- how much checking, from consequence, reversibility, your own uncertainty and whether it is yours to decide -- and the row where the answer is another person.
- [Unit 13: Privacy and data handling](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-13-privacy-data-handling.html): What actually leaves your device, what is retained, and a classification by whose material it is rather than by how secret it feels.
- [Unit 14: Copyright and attribution](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-14-copyright-attribution.html): Three obligations owed separately -- rights, attribution and disclosure -- and honesty about which of them has unsettled answers.
- [Unit 15: Bias, representation and accessibility](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-15-bias-representation-accessibility.html): The method run on its hardest object -- what one controlled comparison establishes, why absence escapes it, and the part with definite answers.
- [Unit 16: The historical bug and historical honesty](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-16-historical-bug.html): An error in a respected published table, the difference between an established fact and a repeated claim, and the course's own mistake with both.
- [Unit 17: Defence, disclosure, and responsible use](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-17-disclosure-defence.html): Say why your answer is right, with evidence someone else can check; say how it was made; know which of the two carries the weight.
- [Unit 18: Alles Lüge! First- and second-order reasoning](https://ai.qyber.dev/edu-ai-literacy-ada-verification/units/unit-18-second-order.html): The closing unit. Where the observer stands when it checks something, why that question is older than AI, and how to sort claims by the evidence actually behind them.

## Optional

- [Glossary](https://ai.qyber.dev/edu-ai-literacy-ada-verification/course/glossary.html): terms used across the course
- [References and source policy](https://ai.qyber.dev/edu-ai-literacy-ada-verification/course/references.html): every cited source and what it may and may not be used to say
- [Handbook](https://ai.qyber.dev/edu-ai-literacy-ada-verification/handbook.pdf): the complete course in one PDF