Where does the "23 Things" idea come from?
The 23 Things format has been around longer than you might expect. It was created in 2006 by Helene Blowers at the Charlotte Mecklenburg Library in North Carolina as a way to encourage librarians to explore Web 2.0 technologies: blogs, wikis, social bookmarking, and other tools that were new and unfamiliar at the time. The idea was simple: give people 23 small, manageable things to try, each with a practical activity, and let discovery do the teaching.
It worked. The format spread quickly, with over 500 adaptations worldwide across libraries, universities, and professional organisations. The University of Edinburgh created an influential version called 23 Things for Digital Knowledge, which won the 2017 LILAC Credo Digital Literacy Award and was openly licensed for reuse. Other organisations adapted the format for their own sectors and audiences, from social services to higher education.
Why 23? The original number was partly practical (enough to cover a good range of topics without feeling overwhelming) and partly inspired by the "23 Enigma," the idea that the number 23 turns up in interesting places. It stuck. Twenty years and hundreds of versions later, the format remains one of the most effective ways to help busy professionals learn new technology through doing rather than just reading.
23 Things AI takes the same approach and applies it to the technology that's changing how we all work: artificial intelligence.
What the programme covers
The 23 Things are organised into phases that build on each other, starting with the fundamentals and gradually introducing more advanced tools and concepts.
Who made this
23 Things AI was created by Rob Stewart, a digital learning consultant based in Scotland.
Rob has spent more than 25 years turning a childhood fascination with digital technology into a career helping people and organisations make the most of digital learning, with more than a decade of experience as a learning and development adviser in the social services sector where he pioneered a new approach to continuous professional development through Open Digital Badges. DigialiTay is where he shares resources, tips, and projects from the world of digital learning.
Rob has spoken at learning and development conferences across the UK, published articles in professional and trade journals, and in 2019 was named Training Journal's L&D Professional of the Year.
23 Things AI is a personal project published on Rob's blog. It's the result of believing that everyone deserves access to practical, honest, jargon-free AI education, and that the 23 Things format remains one of the best ways to deliver it.
How it's written, and how it stays current
Here's something worth knowing about this programme: the content was written collaboratively by Rob Stewart and Claude, Anthropic's AI assistant. Research was gathered with the help of Gemini, Perplexity, and Grok. Every piece of content was reviewed, edited, and fact-checked by Rob before publication.
That might seem like an unusual thing to admit in a programme about AI literacy, but transparency matters here. It's also central to how 23 Things AI stays useful.
The biggest challenge with any resource about AI tools is that things change fast. Tool features change. Pricing tiers shift. New platforms launch while others fade. The traditional approach (write everything, publish, revisit in a year or two) simply doesn't work for this subject. By the time you came back to update, half the content would be out of date.
Instead, 23 Things AI uses AI itself to help monitor and maintain the materials. Rob uses Claude Code to review content against current tool capabilities, flag information that may have drifted out of date, and draft updates that Rob then verifies and approves. It's the same "supervisor mindset" that the programme teaches: AI does the heavy lifting, but a human makes the decisions.
This approach means the programme can stay current in a way that would not be feasible through manual effort alone. Keeping 23 substantial pieces of educational content accurate across a field that shifts monthly would require an unreasonable amount of time for one person working without assistance. With AI in the loop, it becomes manageable. Which is, in a way, what the whole programme is about: using AI as a practical tool that extends what one person can realistically do.
Licence
The content of 23 Things AI is published under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence. You're free to share and adapt the material for any purpose, including commercial use, as long as you give appropriate credit.
This is deliberate. The 23 Things format has always thrived on openness and remixing, from Charlotte Mecklenburg to Edinburgh and beyond. If this resource helps someone create something even better for their own audience, that's the format working exactly as intended.