Skip to content
InfratoApp
← Infrato Academy

Claude Academy videos, in one place

Anthropic’s own short videos on how AI works, when to trust it and how to work with it, gathered so you can watch them here. Nothing loads from YouTube until you press play. Each group links to the tutorial it comes from on Claude Academy.

What happens when you talk to AI?

What happens when you talk to AI? is a five-minute Claude Academy video tutorial on what is going on under the hood in the moment an AI appears to be thinking after you send a message.

Can you trust what AI tells you?

Can you trust what AI tells you? is a five-minute Claude Academy video tutorial on the confident, well-organised and sometimes cited answers an AI gives, and how far to rely on them.

What does AI know about me?

What does AI know about me? is a five-minute Claude Academy video tutorial on what an AI actually knows when you give it details such as medical questions, money worries or emails you are nervous to send.

  • Claude · 2026

    What does AI actually know about you?

    Open on YouTube

The 4 Properties of AI

The 4 Properties of AI is a seven-minute Claude Academy tutorial with four short videos, a quick reference for what makes AI capable in some situations and limited in others: next token prediction, knowledge, working memory and steerability.

  • Claude · 2026

    How AI models use next token prediction

    Open on YouTube

  • Claude · 2026

    Understanding knowledge gaps in AI models

    Open on YouTube

  • Claude · 2026

    The Context Window: How working memory affects generative AI outputs

    Open on YouTube

  • Claude · 2026

    How steerability affects generative AI outputs

    Open on YouTube

Why do AI models hallucinate?

Why do AI models hallucinate? is a five-minute Claude Academy video tutorial on what researchers mean by hallucination, why it may occur, and tactics for spotting it in your own conversations.

What is sycophancy in AI models?

What is sycophancy in AI models? is a six-minute Claude Academy video tutorial on what researchers mean by sycophancy, when it is more likely to show up in conversations, and tactics for steering an AI towards truth.

Why does bias exist in AI models?

Why does bias exist in AI models? is a four-minute Claude Academy video tutorial, presented by an Anthropic researcher, on what bias is, why it may occur, and tactics for spotting bias in your AI conversations.

Writing an AI diligence statement

Writing an AI diligence statement is a fifteen-minute Claude Academy tutorial on creating a transparent, specific disclosure of how AI was used in your work, paired with your commitment to take responsibility for the final output.

  • Anthropic · 2025

    Lesson 10: A closer look at Diligence | AI Fluency: Framework & Foundations Course

    Open on YouTube

A discussion guide for the AI Fluency Index

A discussion guide for the AI Fluency Index is a ten-minute Claude Academy tutorial with an overview video, for teams exploring how to build AI fluency in their organisation alongside the Anthropic Education Report on the index.

What is Claude Managed Agents?

What is Claude Managed Agents? is a four-minute Claude Academy video tutorial on a suite of APIs for building production-ready agents, where you define the tools, environments and success criteria and Claude works until the job is done.

How AI works

How AI works is a Claude Academy collection that starts with three short video tutorials on what happens when you talk to AI, whether to trust its answers and what it knows about you, and goes on to a course on AI capabilities and limitations.