I had a couple of conversations this week that inspired a quick coda to my previous post on using Claude Code for prototyping ideas. The day after I posted it, I spoke with Tom, one of my most thoughtful and creative friends—someone who is about as far away from AI boosterism as it would be possible to get. A decade ago, he had spent several thousand pounds with a web developer building a location-aware mobile website. He asked how much information Claude Code needed to build something similar. I copied his sixty-word WhatsApp, clarified half a dozen questions that Claude asked in response, and it took under two minutes to replicate the core functionality.

Last week, Ethan Mollick posted about creating a functioning game with Claude Code and a one-shot prompt. As a first project with Claude Code, it seemed ambitious but appealing. I started with a single paragraph prompt to develop a short, simple adventure game in the style of the Sierra On-Line games I remember from the late eighties, with an AD&D 2nd Edition vibe. Getting a first, testable version was almost indecently fast. It then took several rounds of iteration to improve it, including a separate workstream to create retro style screens (Claude doesn’t have native image generation). The total active investment of time was about twenty minutes, with Claude Code running in the background while I did other things.

There’s an uncomfortable paradox at the heart of Generative AI adoption: the promise from AI vendors and consultants like me is that the technology will boost productivity. And for the most part, this holds true with my clients. But the gains are unevenly distributed. Dan Hon wrote in his newsletter recently that his friends who use LLMs brilliantly in their coding work do so because they bring very particular skills and experience that don’t translate to the average user—a challenge amplified by what Ethan Mollick calls AI’s ‘jagged frontier’ of uneven capabilities. Scale this up to enterprise level and you hit a bigger problem: AI can create efficiencies for individual, power users, but it doesn’t necessarily deliver the company-wide transformation the C-suite was sold on. You’ve probably seen headlines claiming that most AI pilots fail to deliver measurable P&L impact. Are the promised gains genuinely elusive, or do we need to reframe what we’re looking for?

Recently I’ve been doing some research on the size and scope of the UK book and journal publishing industry, based on data from Companies House. This research is a preparatory step towards building a database of companies, highly relevant for my work as an independent consultant and as policy advisor for the Independent Publishers Guild, which represents over 600 book and journal publishers of all sizes. The other trade body, the Publishers Association lists 168 members at the time of writing, particularly larger, corporate publishers—though there is some overlap between the two organisations’ memberships. The UK publishing sector is world leading, representing over £7 billion of revenue, and it represents the bulk of my client base. So it’s worth understanding.

Last September, BookNet Canada invited me to speak at their Tech Forum webinar on the use of AI in publishing. Looking back at my slides, it’s remarkable how much the world has moved on since then, though the eight principles I set out have held up pretty well.

I was invited to deliver a short webinar today for the Swedish Publishers Association and its international affiliates. It was great to see a very strong turnout and thank you to my hosts for arranging it. I talked about recent trends in AI, how publishers are approaching it, and examples of how AI tools are being used.

Thanks to Graham Bell from EDItEUR for inviting me to speak at the 45th International Supply Chain Seminar in Frankfurt this afternoon, to update the AI presentation I gave last year. Giving an update meant less time on first principles, and it was a really interesting prompt to step back, compare this year’s slides with the 2023 deck and see what had changed. Of course, massive development on the technology side, more legal uncertainty, and greater awareness of the environmental impact of AI. But some points, like the importance of keeping human decision makers in the loop, are perennial.

I gave a short presentation on how publishers are using AI at the Society of Indexers Conference today. It was great to give the publisher perspective ahead of a fascinating presentation from Tanya Izzard on experimenting with LLMs for indexing: the mix of general and specific themes led into a lively discussion.