← back
Why Agentic Systems Need Ontologies — Frank Coyle, UC Berkeley
Takeaway
Give agents explicit domain relationships and inference rules to complement probabilistic reasoning.
Summary
- Ontologies describe domain entities, properties, and relationships, giving probabilistic agents a shared formal representation of an organization's world.
- Neuro-symbolic AI combines neural models with symbolic structures and rules to ground agent behavior.
- Build ontologies top-down with domain experts or bottom-up from customer interactions; reuse vocabularies such as schema.org, FOAF, and Dublin Core.
- RDFS domain/range rules and OWL properties support inference, such as identifying teachers and students or following transitive ancestry relationships.
ontologiesknowledge-graphsneuro-symbolic-ai
Original description
A second refund on the same order. A payout sent to the support desk instead of the buyer. An order status of "probably shipped." These are the kinds of mistakes a probabilistic agent makes and a paragraph of instructions cannot reliably stop. Frank Coyle argues that most agent failures, from brittle tools to fragile handoffs, are symptoms of one missing layer: a formal ontology sitting outside the model as logical guardrails. LLMs reason probabilistically over domains they only half understand, and no amount of prompt engineering closes that gap. His fix is neurosymbolic: probabilistic reasoning inside, logic outside. An ontology is just typed entities, relationships, and constraints, expressed with old and boring standards like RDFS and OWL, that let you say a payment status must be one of three values, that a customer and a support rep are different things, that an order can only be refunded once. Wrap a Claude tool use loop with a validator: when the model proposes a tool call, check its types with Pydantic and its results against the ontology, and only then let it act. The catches that are painful to write in English become a few lines of logic. Speaker info: https://x.com/coyle_frankp / frank-coyle https://www.frank-coyle.ai/ Timestamps: 0:00 - Intro and an educator's philosophy 2:21 - Two lineages: agents and ontologies 4:04 - Neurosymbolic AI: guardrails around a probabilistic model 5:23 - What an ontology actually is 6:14 - Building one, and the expert systems era 7:55 - Reusing existing taxonomies 9:12 - RDFS and OWL: inference and constraints 12:12 - Agents, loops, and how they break 14:22 - A Claude tool use loop with an ontology validator 17:47 - Pydantic at the door, ontology at the ledger 18:52 - The errors an ontology catches that English cannot