The Jellyfish and the Flatworm. A Story About AI Strategy

Is your organization more like a jellyfish or a flatworm?

The author's "Jellyfish and Flatworm story has been remarkably effective at helping ... [executives] visualize the impact of AI on their customers, their products, and their employees... this story is about why Knowledge Representation (KR) must be the core of any cost-effective long-term AI strategy... how LLMs..., Knowledge Graphs (KGs), and Reference Frames (RFs) are moving us closer to general AI and how building hybrids of these three knowledge representation strategies is the best path".

Evolution: a tale of 2 paths

As the storytitle implies, the metaphor stems from the natural world, specifically the evolution of intelligence:

• "Jellyfish live in the open ocean, far away from complex structures... only needs simple rules to navigate its environment... they are not hunters. They depend on fish drifting into their tentacles". Powering an energy-intensive brain is impossible: "they need to keep things simple."
• Not so for flatworms on the ocean floor, which "to seek their prey and avoid predators... started to develop muscles to help them move around... cells on their skin that could process complex signals such as light, temperature, and smell". For these first hunters, "Knowing both what rules to follow and when to follow them became more complex... Flatworms started to centralize where these rules were executed... near their front-facing sensors" - what we now call 'heads', containing their "centralized nervous system ... the brains of animals that move about in the world... [because] movement makes executing rules complicated... intelligence and models of our world are tied to motion and, importantly, maps and structure."

Evolution pushed animals' brains, or "predictive organs", to become capable of building more and more complex models of the outside world.

LLMs, Knowledge Graphs & Reference frames

LLMs do not do this, and were "never designed to store accurate models of the real world... only a shadow of how we communicate about the world using language... [hence LLMs'] severe limitations with precision, reproducibility, truthfulness, performance, and explainability".

Knowledge graphs, on the other hand, "are the closest thing we have today to modeling our world in computers.

Our brains, finally, use Reference Frames to store data. These "evolved from building maps of our world... Unlike an LLM, their knowledge can be continually updated. And just like scale-out distributed knowledge graphs, their processing is done in parallel."

Company strategy

Unless your company is a jellyfish (simple product, few clients, no competitors) you are a flatworm, selling many products to many customers, for each product many competitors. "you need a complex model of your world... that include structure, relationships, precision, explainability, and the ability to add new complexity", if you are to be able to "simulate the impact of a price increase... predict the impact of a new marketing campaign... explain why sales of some items are dropping ... If you can’t do this ... your model of the real world is [probably] too simple and too flat without structure. You might need to invest in using a combination of knowledge graphs and LLMs to accelerate your ability to predict the future. "