Structured data is the glue for building interactions between AI-mediated experiences—between system and people, and between system and system. Great to see the tech evolving to support new and inventive (and radically adaptive) experiences.
Founder of Big Medium, a digital agency that helps complex organizations design for what’s next. We build design systems, craft exceptional online experiences, and transform digital organizations.
While there's tons of focus right now on how we talk directly with LLMs, one of the things that I'm most excited about is how they can talk to other systems. When AI-mediated experiences are interoperable, exciting things happen. Reliable structured data is at the root of it—It's what makes agents go and what enables radically adaptive UI. So I was unreasonably excited to see OpenAI announce the addition of new discipline to the model to help ensure the reliability of the structured data it returns. Most of us experience OpenAI’s GPT models as a chat interface, and that’s certainly the interaction of the moment. But LLMs are fluent in lots of languages—not just English or Chinese or Spanish, but JSON, SVG, Python, etc. One of their underappreciated talents is to move fluidly between different representations of ideas and concepts. Here specifically, they can translate messy English into structured JSON. What this means for product designers: Moving nimbly between structured and unstructured data is what enables LLMs to suggest and deliver interface elements, to make function requests to other systems, to act on your behalf, or to share information in new ways. The interoperable ability of Sentient Design systems to "speak machine" is the stuff that will animate the next generation of interaction design. Alas, as in all things LLM, the models sometimes drift a bit from the specific ask—the JSON they come back with isn’t always what we asked for. This latest update is a promising direction for helping us get disciplined responses when we need it—so that Sentient Design experiences can reliably communicate with other systems.