At FactSet’s FOCUS 2026 conference in Austin, Texas, the buy side’s structural pressures were impossible to miss. Buy-side firms are wrestling with margin compression, fragmented legacy technology, and complex regulatory landscapes. When you throw in the race to operationalize AI and a volatile geopolitical environment, you get a pretty compelling agenda.

AI is moving from experimentation to execution

KPMG found that only 32% of financial services companies are generating returns from AI at scale and 67% have not yet reached this stage.ⁱSessions like “Leveraging AI in an Open and Flexible Programmatic Environment” cut to the real question: how firms are actually operationalizing AI.ⁱⁱ Conversations addressed what it takes to build AI-ready data ecosystems at scale, where firms are making genuine progress, and most importantly where they are stalling.

The consensus held AI as a workforce multiplier rather than a headcount reducer. Our recently launched Arcesium Intelligence is designed to help firms thoughtfully deploy agentic AI across investment operations and data management workflows through a model-agnostic orchestration layer. Institutions want and need complete control over which models they use, how much human oversight applies, and how the full chain is audited.

The race to AI is a big reason the industry’s total cost base rose to $167 billion in 2024, a 7% increase.ⁱⁱⁱ An uptick in technology spending has firms searching for ways to demonstrate ROI from AI initiatives. AI-ready data infrastructure was another major theme, since AI’s only as effective as the data foundation on which it lives.

1. Data infrastructure is the new competitive moat

We’ve already been urging the institutional investment ecosystem to start thinking of data as a strategic asset. Poor data quality directly degrades AI model performance and produces unreliable insights. Several sessions explored the big data problem, including “Breaking Data Silos: Moving from On-Prem to the Cloud” and our own “Operational Alpha: Rethinking Data Management.” The broader message was that firms with clean, connected, and semantically structured data are best positioned to benefit from AI.

Discussions detailed how organizations are re-architecting market data ecosystems away from legacy systems toward cloud-native, globally scalable platforms. The urgency is straightforward: Asset managers and owners are inundated with a growing volume of digital information. The trouble is that many firms lack the ability to fully unlock the power of their data without technology to harmonize and govern it. Those who can use all of their data, when they need to use it, have a competitive edge.

A mere 18% of front-office teams report being able to get timely access to data without requiring manual intervention from other departments.^iv The increasing ubiquity of alternative asset classes is one of the “great complicators” of data management in finance.

2. Public and private markets are converging

Multiple sessions, like “Data as Alpha: Integrating Public and Private Market Intelligence,” tackled how asset managers and owners are rethinking portfolio construction as private credit and alternatives, and how illiquid assets now sit alongside traditional equity and fixed income in meaningful ways. The concept of the 60/40 portfolio is waning and moving toward a 50/30/20, with private markets siphoning share away from equities and fixed income. This transformation will likely accelerate, and one speaker predicted that private markets may grow from roughly $15 trillion today to possibly $60 trillion over the next five to ten years, given the democratization of access to these asset classes within the retail and retirement space.

This growth and the need for consistent and accurate reporting is increasing the importance of a total portfolio view across public and private markets. Institutional limited partners (LPs) now expect real-time visibility into performance, fees, and portfolio metrics. Firms that fortify their data foundations can transform the investor experience and their own workflows.

In the session “Navigating Data Complexities in Investment Operations,” thought leaders from Northern Trust and Wellington Management joined FactSet’s Arun Veluswamy to discuss solutions to the complexity of data governance and private credit. The public-private asset class convergence has changed our data management paradigm. In addition to making alternative markets transparent, we also need to see the sum and product of a complicated mix of inherently opaque markets as they intertwine with public markets. To know the total fund portfolio view, systems must work on a foundation of consolidated, standardized data.

3. Why operational alpha is redefining investment operations

Sessions like “From Reports to Relationships: The Digital Transformation of Client Reporting” examined how middle- and back-office roles are becoming a source of competitive advantage. These roles are changing from a historical data intake and reporting model into a forward-looking analytics approach that provides an efficient operational environment for investment decision makers to focus on their investment performance.

Moreover, operations are delivering greater value by doing things like optimizing cash allocations, identifying opportunities to claw back excess unencumbered cash, and eliminating information asymmetry in securities borrowing and lending. AI is poised to take this to another level, freeing middle- and back-office professionals to spend more time on higher-value work.

Will AI live up to front-office alpha ambitions?

How various institutional investment firms approach the next phase of AI investment operations transformation will define the coming years. Mercer’s 2026 AI in Asset Management Survey revealed that managers are keeping the technology at arm’s length when it comes to the decisions that actually move markets; and in at least one of their strategy’s investment processes, 27% are at pilot or proof-of-concept, and only 18% report no integration yet. Time will tell if AI lives up to front-office alpha generation ambitions.^v

Firms that are walking a line between enthusiastic AI experimentation and keeping AI at arm’s length should guard against both becoming overly cautious and aggressively haphazard. Regardless of firms’ current place on the AI adoption curve, they should build from the data foundation ground up, with human-led clarity and tailored to the unique needs of the firm.

You can see additional highlights of FactSet Focus 2026 here.