Semantic Data New York 2026
Program Update: First Sessions Announced
This year we’re introducing a number of interactive sessions where attendees can ask questions and explore key issues with others in the room. Designed to complement the wider program, they create space for open exchange and shared thinking.
Semantics in the Age of AI: Questions Practitioners Must Answer in 2026
Context and challenge
As AI systems become the primary way people and applications access information, the semantic layer (ontologies, knowledge graphs, and contextual metadata) has moved from nice‑to‑have to critical infrastructure. Yet many organisations still struggle to decide when to invest in graphs, how to manage change, and how to align semantic work with data and AI teams.
Meeting the challenge
This collective roundtable discussion focuses on answers to questions that increasingly determine whether semantic initiatives deliver real value or remain science projects.
Join the conversation, share your experiences and learn what others are doing.
Moderator: Dr. Robert Sanderson, Senior Director for Digital Cultural Heritage, Yale University
LLM, Semantic Classification, or Both? Designing the Right Approach for Enterprise Data
Organizations are increasingly evaluating large language models (LLMs) for content classification and tagging. While LLMs offer flexibility and can accelerate semantic model development, they also introduce considerations around consistency, explainability, governance, and cost.
This session compares LLM-driven classification, semantic classification, and hybrid approaches that combine the strengths of both. Through practical demonstrations and real-world examples, attendees will see how AI-assisted modeling can accelerate semantic model creation, while semantic classification delivers consistent, repeatable tagging across large volumes of enterprise content.
There will also be an examination of an often-overlooked architectural consideration: cost predictability. As document volumes grow, content is reprocessed to meet new business requirements, and AI licensing models evolve, the cost of token-based classification can become increasingly difficult to forecast.
Lance Thieshen, Senior Principal Sales Engineer, Progress Software
Semantics, the ‘DNA’ of Biotechnology Data
Context
- The biotechnology industry has embraced semantic data in the form of taxonomies and ontologies for several decades. Examples include the Gene Ontology, dating to 1998, and the Cell Ontology, which was initially released in 2004.
- While the use of semantic data on the research and development side of biotechnology is well established, changes in the semantic industry on the business side are evolving more rapidly.
- Many biotechnology companies face the same challenges as other industries when it comes to data:
- Siloed systems with standalone metadata structures.
- Using the wrong systems to manage data and assets.
- Inadequately managed data for use in reporting and analytics.
With the sudden growth and adoption of artificial intelligence, ontologies, knowledge graphs, and semantic layers are being viewed with renewed interest in biotechnology business operations.
Covering
- How semantics is used in the biotechnology industry.
- How artificial intelligence is driving semantic adoption to meet marketing use cases.
- The general relevance of what the biotech industry is learning and doing for other managers of semantic data.
Ahren Lehnert, Senior Taxonomist, Genentech
From MARC to Meaning: Building and Using a Knowledge Graph at Library Scale
Libraries have cataloged the world's published record for over a century, but that description sits as text inside records rather than as identifiable things. We have spent the past several years rebuilding it as a knowledge graph. WorldCat Entities publishes descriptions of people, events, places, organizations, and works as persistent, resolvable URIs. The Dewey Decimal Classification is published on the same basis, which turns a familiar classification scheme into a machine-navigable hierarchy of human knowledge.
These identifiers have been integrated across OCLC data to improve our products and services and are also widely used by libraries and other cultural heritage organizations around the world.
Two decisions mattered most in the effort to build out our knowledge graph.
- First, we centralized entity management so that two libraries describing the same person land on the same URI.
- Second, we never asked anyone to adopt a new workflow: identifiers are written into records automatically as catalogers do the job they already do, which is why adoption has grown steadily rather than stalling.
An important lesson we learned that can help every organisation working to build a graph is that a graph delivers its value at the point of linkage, so getting identifiers into the systems people already use is critical to success and adoption.
Jeff Mixter, Director of Global Metadata and Innovation, OCLC
The Semantic Museum: Data and Disciplinarity
Context
The Getty, based in Los Angeles, is a unique cultural institution, combining an art museum, library, archive, foundation, and conservation science program. Shared across all of these is a deep commitment to creating, managing, and sharing knowledge of the visual arts—as well as building the cultural heritage field's capacity to work with this information.
What we have done
Over the past decade, Getty has implemented the infrastructure and practices needed to unify its archival, museum, research, and imaging data by applying semantic technologies. Some of the work has been technical–building models, setting up infrastructure, developing taxonomies, and leveraging standards. But far more important are the social processes that enable the technology: building the internal capacity needed to execute this work and the willingness to look beyond disciplinary practice and consider what it really means to connect knowledge across an institution's many centers of excellence.
What we have learned
- The challenges involved in bringing together all the different strands of knowledge together.
- How you prevent the technology getting ahead of the capacity of the humans generating the knowledge.
- How you find that sweet spot between representing the world's true complexity and building something impenetrable to the very audience you hope to serve.
David Newbury, Senior director, Public technologies, Getty
More to follow….
Click here to register