Quantitative investment teams are placing increasing emphasis on sector- and industry-specific datasets as artificial intelligence becomes more deeply embedded in research workflows, according to a new survey conducted by Bloomberg.
The study, based on responses from more than 150 quants, research analysts and data scientists across EMEA and North America, suggests the industry is reaching an inflection point in AI adoption – with progress now shaped less by experimentation and more by data readiness.
While traditional machine learning techniques are now well established within quantitative strategies, the survey found that 54% of respondents have yet to begin using generative AI in live research workflows. Bloomberg said this reflects ongoing challenges around structuring and contextualising data, which remains a prerequisite for effective generative AI and agent-based research models.
As investment strategies become increasingly specialised, demand for sector-specific and domain-driven data is growing. Nearly three-quarters of respondents (72%) said they are looking to onboard more granular datasets, including company-level key performance indicators, pharmaceutical pipeline data and semiconductor segment revenue breakdowns.
The survey also highlights how AI is currently being applied across quantitative research. The most common use case cited was stock selection, identified by 48% of respondents, followed by content summarisation (21%) and thematic analysis (13%). The majority of participants indicated that their strategies are primarily focused on equities.
According to Bloomberg, the shift toward deeper, industry-level datasets reflects a broader move toward more context-rich inputs for alpha generation, alongside traditional market data such as tick history and measures of investor expectations.
Angana Jacob, Global Head of Research Data at Bloomberg Enterprise Data, said the findings illustrate a change in how firms approach AI integration. “Progress with AI is increasingly shaped by data readiness rather than experimentation alone,” she said. “Firms are prioritising sector-specific datasets that provide deeper company and industry context, supporting more sophisticated quantitative workflows.”
The survey builds on Bloomberg’s 2025 Research Data study, which identified data sourcing, integration and governance as key challenges facing quantitative research teams. This year’s findings suggest those structural issues remain central as firms prepare for wider adoption of generative AI across investment research.
