Short course in United States
Agentic AI Foundations: Business Applications and Risks
A four-week instructor-led introduction from Harvard SEAS to how AI agents work, where they fit in business, and what can go wrong.
Our take
A tidy, affordable primer taught by Hanspeter Pfister that covers both mechanics (LLMs, reasoning, tool use) and risks such as hallucination and data exposure. It is conceptual rather than hands-on, so do not expect to build agents.
Best for: Managers and professionals who need a non-technical grounding in agentic AI before making adoption decisions.
Based on the course page, syllabus and pricing, not on completing the course. How we rank
Good to know
Runs 2-30 December 2026 on Harvard Online; part of the AI Leadership Series.
What you'll study
- Agentic AI
- Large language models
- Task decomposition
- Tool integration
- AI risk and governance
Before you start
- None listed; introductory level.