Governance as the cornerstone of AI success in the enterprise
Enterprise AI is moving from single prompts to long-running systems. That makes governance an architectural job, enforced at runtime, not a policy document.
Insights & Publications
Thinking that informs the work.
Independent perspectives on enterprise AI, alongside selected scientific publications by our partners. Explore the decisions, methods, and ideas behind the practice.
Need this examined for your own decision? Commission research
14 results
From Generative AI to Decision Intelligence
Generative AI made content cheap. The enterprise prize is different: better decisions, made faster, at scale, with accountability intact.
AI Is Becoming a Portfolio, Not a Project
Treating AI as a set of projects produces a museum of pilots. Treating it as a managed portfolio produces compounding advantage.
The Industrialization Gap in Enterprise AI
The distance between a working pilot and a production capability is where most enterprise AI value is quietly lost.
Build, Buy, or Partner? A Framework
The most consequential AI decisions are not about models. They are about what to own, what to rent, and what to build together.
From AI Models to Decision Systems
A model produces a prediction. A decision system produces an action, an owner, and an audit trail. Enterprises need the second.
Why Agentic AI Is an Organizational-Architecture Problem
Agents fail in production for the same reasons organizations do, unclear authority, weak escalation, and no accountability.
What Boards Need to Understand About AI
Boards do not need to understand transformers. They need to understand exposure, investment, and the questions to put to management.
Why Data Maturity Is Still AI Maturity
The generative era did not repeal the old truth: an organization's AI ceiling is set by its data foundation.
AI Sovereignty and the Emerging-Markets Opportunity
The leapfrog thesis is real, but it is not automatic. It depends on strategy fit to local infrastructure, economics, and institutions.
Value Is Moving From Models to Applications
As foundation models commoditize, durable advantage is migrating to the application layer with the decisions, data, and workflows around them.
Time-resolved topological data analysis of market instabilities
An investigation of topological methods for identifying changes in financial time series. The study examines credit default swap data and explores signals associated with approaching market instability.
Topological Data Analysis of Financial Time Series: Landscapes of Crashes
The authors examine the evolution of US stock-market time series around major financial crises, using persistent homology to study patterns in the data and potential early warning signals.
IBM InfoSphere Streams for scalable, real-time, intelligent transportation services
A prototype demonstrates how stream processing can combine vehicle GPS and road-network information to produce real-time transportation insights and routing information.
Bring clarity to your next AI decision.
Tell us what your organization needs to accomplish. We’ll start with the decision in front of you.