Long before AI became the dominant technology conversation in investment management, we held a simple view at Lightkeeper: the quality of your portfolio analytics is a direct function of the data underneath them. AI does not change that reality. If anything, it makes it more important. Clean, validated, consolidated data is not a precondition you get to skip. It is the required work. And if that work has not been done, the sophistication of the AI layer on top is largely irrelevant.
We are not the only ones saying it anymore. The Hedgeweek 2026 Hedge Fund Technology Report, produced in collaboration with SS&C and based on a survey of over 100 hedge fund managers, has put hard numbers behind what many in the industry already sense. The picture it paints is one of a market accelerating into AI investment while quietly deferring the foundational work that would make that investment pay off.
The Spending Picture
The report confirms that technology budgets are moving. 61% of funds increased their technology spend over the past 12 months. AI and machine learning top the stated priority list at 32%, with trading and portfolio management systems at 24% and data infrastructure a distant third at 22%.
These spending priorities are telling. Not because AI should not be a priority, but because of what they reveal about the order in which firms are prioritizing technology investments. Most investment managers would agree that the quality of their data ultimately determines the quality of their reporting, analytics, and decision-making. Yet those same managers are prioritizing AI over the data infrastructure that would make it work.
The era of broad experimentation, as the report notes, is over. What gets approved now is anything that consolidates workflows, not expands them. Firms are becoming less interested in adding another tool and more interested in reducing operational complexity. But consolidation only delivers if the foundation being consolidated is worth building on.
The Gap No One Wants to Talk About
The report states that a quarter of managers in the survey describe their technology infrastructure as behind where it needs to be. Of that group, two-thirds are prioritizing AI and trading systems. Only 7% are prioritizing data infrastructure.
This might be the most interesting insight in the report. The firms that openly admit their infrastructure is not ready are still prioritizing AI over the foundation it requires. The report does not soften this finding. One of the research contributors put it plainly: putting AI on top of broken infrastructure produces “a Ferrari engine in a car with bad brakes.” The unglamorous work has to come first.
The pressure to show visible AI progress often wins out over the less visible work of fixing what sits underneath. Most of these firms are not confused about what needs to happen. They know their data environment needs work. The problem is that AI initiatives are easier to showcase than the operational work required to support them.
Meanwhile, AI is delivering measurable results in a narrower slice of use cases than the investment levels would suggest. Document processing is the only application showing broad adoption and tangible impact, cited by 42% of respondents. Thirty percent of funds report no measurable AI impact at all. Technology is rarely the limiting factor. The quality of the underlying data and workflows still determines how much value firms actually realize.
What We See on the Ground
We see the downstream consequences of this pattern regularly. When a fund comes to Lightkeeper with data spread across disconnected systems, positions that do not reconcile cleanly, and reporting that still depends on manual aggregation, the conversation about AI has to wait. Not because we do not want to have it, but because the honest answer to ‘how do we get more out of AI’ is almost always ‘fix what is underneath it first.’
The funds that have done that work are the ones asking better questions and getting faster, more reliable answers from every tool they use, AI included. The advantage comes less from the AI itself than from the groundwork that was completed years earlier.
What Good Sequencing Actually Looks Like
The mid-sized funds the report identifies as doing the foundational work while others chase headlines, are not behind. They are, in the study’s own framing, sequencing. There is a real difference between a firm still untangling its data environment and one that has already deployed AI on top of a broken foundation and is now discovering the limits of that decision.
Good sequencing does not mean delaying AI indefinitely. It means putting the prerequisites in place first. The firms seeing the strongest results today are typically the ones that invested in data quality, workflow consistency, and operational discipline. AI becomes far more valuable when those pieces are already in place.
The Question Worth Asking
The Hedgeweek report ends with six words that cut through most of the noise in this conversation: process clarity comes first, then tooling. For any fund evaluating its AI roadmap in 2026, the most useful question is not which AI tool to buy. It is whether the data environment underneath is structured well enough to make that tool deliver.
Achieving that clarity doesn’t mean putting innovation on hold. At Lightkeeper, we spent years building a validated, consolidated data foundation because we knew it was the only way portfolio analytics could truly deliver. Today, it is exactly what our AI capabilities are built on top of. The funds that sequence this well are the ones that will look back and realize the AI advantage started long before the AI.
