Official documentation, open courses, frameworks, and community labs—reviewed by IHA, mapped to real skills, and always linked to the original creator.
We explain why a resource belongs and when it was last checked.
Stale or superseded material is refreshed or archived—not left to mislead learners.
Watching is valuable, but credentials require original work and reviewable evidence.
10 reviewed resources
University of Helsinki and MinnaLearn
A non-technical introduction to what AI can do, how common methods work, and where human judgment still matters.
Why IHA recommends it
A durable conceptual foundation before learners choose tools or automate work.
OpenAI
Official learning material and events for applying OpenAI products across work, education, and community settings.
Why IHA recommends it
The right primary source when a learner needs product-specific OpenAI instruction.
Anthropic
Official notebooks covering API fundamentals, prompt engineering, evaluations, tool use, and retrieval workflows.
Why IHA recommends it
A strong vendor-authored technical path that learners can compare with other model providers.
OpenAI
Official examples and guides for building, evaluating, and operating applications with OpenAI models.
Why IHA recommends it
Practical reference implementations from the product maker, best used alongside independent evaluation.
Hugging Face
Hands-on instruction in transformers, model use, fine-tuning, datasets, tokenizers, and sharing open models.
Why IHA recommends it
A credible route beyond closed model interfaces for learners ready for technical depth.
GitHub
Short project-based courses that teach GitHub by guiding learners through real repository actions.
Why IHA recommends it
The interaction model is excellent: small authentic tasks, visible artifacts, and automated feedback.
National Institute of Standards and Technology
A voluntary framework for governing, mapping, measuring, and managing risks across the AI lifecycle.
Why IHA recommends it
A vendor-neutral backbone for translating responsible-AI principles into operating controls.
OWASP GenAI Security Project
A practical taxonomy of consequential security risks in applications that use large language models.
Why IHA recommends it
It turns abstract safety concerns into failures learners can test and design against.
U.S. Federal Trade Commission
Plain-language business guidance on substantiating AI claims and avoiding deceptive performance promises.
Why IHA recommends it
A short, authoritative test for whether an AI-enabled offer is honest and supportable.
U.S. Small Business Administration
A practical overview of direct research, existing sources, competitive analysis, and market questions.
Why IHA recommends it
It reinforces that business decisions require evidence beyond model-generated market summaries.