LLMs will represent a step change in tool support for end-user programming: the ability of normal people to fully harness the general power of computers without resorting to the complexity of normal programming. Until now, that vision has been bottlenecked on turning fuzzy informal intent into formal, executable code; now that bottleneck is rapidly opening up thanks to LLMs.

If this hypothesis indeed comes true, we might start to see some surprising changes in the way people use software: One-off scripts: Normal computer users have their AI create and execute scripts dozens of times a day, to perform tasks like data analysis, video editing, or automating tedious tasks. One-off GUIs: People use AI to create entire GUI applications just for performing a single specific task—containing just the features they need, no bloat. Build don’t buy: Businesses develop more software in-house that meets their custom needs, rather than buying SaaS off the shelf, since it’s now cheaper to get software tailored to the use case. Modding/extensions: Consumers and businesses demand the ability to extend and mod their existing software, since it’s now easier to specify a new feature or a tweak to match a user’s workflow. Recombination: Take the best parts of the different applications you like best, and create a new hybrid that composes them together.

Chat will never feel like driving a car, no matter how good the bot is. In their 1986 book Understanding Computers and Cognition, Terry Winograd and Fernando Flores elaborate on this point: In driving a car, the control interaction is normally transparent. You do not think “How far should I turn the steering wheel to go around that curve?” In fact, you are not even aware (unless something intrudes) of using a steering wheel…The long evolution of the design of automobiles has led to this readiness-to-hand. It is not achieved by having a car communicate like a person, but by providing the right coupling between the driver and action in the relevant domain (motion down the road).

Think about how a spreadsheet works. If you have a financial model in a spreadsheet, you can try changing a number in a cell to assess a scenario—this is the inner loop of direct manipulation at work. But, you can also edit the formulas! A spreadsheet isn’t just an “app” focused on a specific task; it’s closer to a general computational medium which lets you flexibly express many kinds of tasks. The “platform developers"—the creators of the spreadsheet—have given you a set of general primitives that can be used to make many tools. We might draw the double loop of the spreadsheet interaction like this. You can edit numbers in the spreadsheet, but you can also edit formulas, which edits the tool

what if you had an LLM play the role of the local developer? That is, the user mainly drives the creation of the spreadsheet, but asks for technical help with some of the formulas when needed? The LLM wouldn’t just create an entire solution, it would also teach the user how to create the solution themselves next time.

This picture shows a world that I find pretty compelling. There’s an inner interaction loop that takes advantage of the full power of direct manipulation. There’s an outer loop where the user can also more deeply edit their tools within an open-ended medium. They can get AI support for making tool edits, and grow their own capacity to work in the medium. Over time, they can learn things like the basics of formulas, or how a VLOOKUP works. This structural knowledge helps the user think of possible use cases for the tool, and also helps them audit the output from the LLMs. In a ChatGPT world, the user is left entirely dependent on the AI, without any understanding of its inner mechanism. In a computational medium with AI as assistant, the user’s reliance on the AI gently decreases over time as they become more comfortable in the medium.