Reading librarySources checked 16 November 202627 sources
The best and most recent reads on AI
A short library, from the basics to the advanced. Every source is checked and dated.
Basics
For business owners and employees. No technical background needed.
- Inside the AI Index: 12 Takeaways from the 2026 ReportStanford's yearly scorecard on AI. This recap pulls out the numbers that matter most for a non-technical reader.Read
- The state of AI in 2026: On the road to ROIA global survey of companies using AI. Shows how many actually see it show up in their bottom line.Read the report
- Large Firms With at Least 20 Employees Biggest AI UsersOfficial US data on which size of business actually uses AI today, and which ones lag behind.Read
- Does Using Artificial Intelligence Save Time at Work?Asks workers directly if AI gave them back time on the job, with a real number behind the answer.Read
- Empowering SMEs in the Age of AIA policy report on how small and mid-size businesses take up AI. Written for policymakers, readable for owners.Read the report
- AI For EveryoneA short, free course built for people with no technical background. A good first stop before anything else on this page.Take the course
- Choosing to Stay HumanA clear, readable take on what AI changes for how people learn and work, from a researcher who studies this closely.Read
- Davos: Here's What to Know About Jobs and Skills TransformationA plain look at how AI is expected to reshape jobs over the next few years.Read
- AI Agents and GenAI: What's the Difference? This Expert ExplainsA short video that explains, in plain words, the difference between a chatbot and an AI agent that acts on your behalf.Watch
Practitioner
For people who already use AI at work and want to use it well.
- Management as AI SuperpowerA framework for deciding when to hand a task to AI versus keep it yourself, with real numbers on how often AI already matches experts.Read
- A Guide to Which AI to Use in the Agentic EraExplains why the same AI model can feel smart in one tool and outdated in another. The app around the model matters as much as the model.Read
- 2025: The Year in LLMsA long, honest year-in-review from someone who tests AI tools daily. Good for catching up fast on what actually changed.Read
- Prompt EngineeringThe official OpenAI guide to getting better answers out of its models, with concrete techniques, not just tips.Read
- AI Fluency: Framework and FoundationsA free, self-paced course for anyone who already uses AI at work and wants to use it with more judgment.Take the course
- Generative AI for EveryoneA short course that goes one level deeper than the basics, aimed at people applying AI inside a real job or business.Take the course
- 2026 Work Trend Index: Agents, Human Agency, and OpportunitySurvey data from thousands of AI users on what actually frees up their time, and what still needs a human check.Read the report
- Ethan Mollick on Staying Human at WorkA conversation on which human judgment still matters most as AI takes on more routine work.Watch
Advanced
For engineers and technical leaders building with AI.
- Building Effective AI AgentsThe reference most engineering teams still point to first when designing an AI agent. Lays out patterns, not hype.Read
- Writing Effective Tools for AI AgentsPractical guidance on designing the tools an agent calls, since a badly designed tool breaks an agent faster than a bad prompt.Read
- The Lethal Trifecta for AI AgentsNames the exact combination of risks that makes an AI agent exploitable: private data, untrusted content, and a way to send data out.Read
- AI Evals: Everything You Need to KnowA working reference on how to actually test whether an AI system works, written by two people who do this for a living.Read
- Product Evals in Three Simple StepsA short, concrete method for comparing two versions of an AI feature without fooling yourself about which one is better.Read
- Software Is Changing (Again)A keynote on how writing software is shifting again, this time toward prompting models instead of only writing code.Watch
- LLM CourseA free, hands-on course covering how language models actually work under the hood, for engineers who want to go deeper than APIs.Take the course
- Evaluating AI AgentsA short course on testing agents specifically, since the usual ways of testing software do not fully cover an agent's behavior.Take the course
- OWASP Top 10 for LLM Applications 2025The closest thing to a standard checklist of security risks in AI applications, from the group behind the original OWASP Top 10.Read the paper
- Agentic Context Engineering: Evolving Contexts for Self-Improving Language ModelsA research paper on letting an AI system improve its own instructions over time, instead of a person rewriting them by hand.Read the paper