Publication
Panel on AI for Future Databases: A New Beginning or a Boulevard of Broken Dreams?
Carsten Binnig; Danica Porobic
In: Volker Markl; Joseph M. Hellerstein; Azza Abouzied (Hrsg.). Companion of the 2025 International Conference on Management of Data, SIGMOD/PODS 2025, Berlin, Germany, June 22-27, 2025. ACM SIGMOD International Conference on Management of Data (SIGMOD), Pages 1-1, ACM, 2025.
Abstract
AI has opened new directions in database research, from learned
components replacing traditional internals to large language mod-
els(LLMs), enabling a new generation of database systems that allow
querying data beyond tables. Yet, adoption in commercial databases
has been incremental rather than a fundamental rethinking of mod-
ern data system stacks. In this panel, we bring together experts from
academia and industry to discuss the tension between potential
and reality in how AI shapes real-world database products.
We will explore questions such as: What should an AI-ready
database stack look like: incremental evolution or radical departure?
What prevents AI from replacing traditional components like query
optimizers, cost models, and indexes? What does it take for LLM-
based innovations to move beyond impressive demos? Can we
use LLMs for more than Text-to-SQL and LLM-UDFs? By tackling
these questions, this panel will challenge assumptions in research,
examine the role of AI in future databases, and ask the following:
Is AI the key to overcoming core limitations and will thus enable a
new generation of database systems, or maybe AI is just another
boulevard of broken (database) dreams?