In this edition, our Talent Intelligence team investigates how artificial intelligence (AI) is reshaping commodity trading, from the skills in demand and barriers to adoption to future sources of competitive advantage.
To access the full report, please complete the form at the end of this article.
Routine But Uneven
Artificial intelligence is now a routine part of how many commodity trading organisations operate. What began as experimentation has, in several areas, translated into tangible commercial outcomes.
At the same time, progress has been uneven, with value emerging more quickly in some parts of businesses than others. To understand what is driving this divergence and what it means for future performance and skillsets, HC Talent Intelligence, in partnership with FT Longitude, surveyed 131 senior executives across trading houses, asset‑backed organisations and financial institutions globally.
Where is AI Delivering Value Today?
As the chart above shows, survey respondents report that AI is already delivering measurable commercial value across commodity trading markets, particularly in risk management, price forecasting and market analytics – broadly reflected in the capabilities commodities organisations are searching for. These functions consistently stand out as areas where AI is having the greatest impact, reflecting their reliance on structured data and their proximity to core decision-making.
In practice, AI is not yet transforming trading outcomes end-to-end. Instead, its contribution is concentrated in analytically intensive decision-support activities, where it can enhance speed and precision without displacing human judgement. This pattern helps explain both the pace of early adoption and its limits: value has emerged first where data quality is highest, feedback loops are clear, and integration into existing workflows is most straightforward.
Overall, the findings suggest that AI has reached a point of commercial relevance, but not comprehensive transformation. Its impact today is meaningful but selective, laying the groundwork for broader change rather than delivering it in full.
Why is Scaling Still a Challenge?
As organisations look to extend these gains, a consistent constraint becomes apparent. As our heatmap above shows - across the survey, data quality is cited as the single most important barrier to scaling AI, ranked ahead of systems integration, governance, talent or questions around return on investment.
This emphasis on data is revealing. It suggests that the challenge is no longer primarily about access to advanced tools or technical capability, but about the foundations required to deploy AI reliably and at scale. Where data is fragmented, poorly governed or difficult to integrate across systems, AI initiatives struggle to move beyond isolated use cases.
Importantly, respondents indicate that these constraints evolve as adoption progresses. Earlier in the journey, organisations tend to focus on questions of capability and confidence. As deployment advances, attention shifts toward integration, governance and operational reliability. In effect, barriers change character rather than disappear altogether, reinforcing the idea that scaling AI is as much an organisational and data challenge as it is a technological one.
From a talent perspective, scaling AI requires traders and risk specialists able to consume model outputs, technologists who understand trading workflows, and data specialists who ensure models are trusted – profiles that remain in short supply within many trading‑led organisations.
From a talent perspective, scaling AI requires traders and risk specialists able to consume model outputs, technologists who understand trading workflows, and data specialists who ensure models are trusted – profiles that remain in short supply within many trading‑led organisations.
Where Will Competitive Advantage be Won?
Looking ahead, respondents increasingly frame AI as a means of strengthening existing competitive advantages rather than redefining business models. Superior proprietary data, faster decision cycles and more accurate risk pricing emerge as the most frequently cited sources of future differentiation.
Notably, less emphasis is placed on highly visible or fully automated outcomes. Instead, respondents point to advantages that are incremental, cumulative and difficult to replicate, rooted in internal data assets, workflow integration and decision discipline. This highlights a growing recognition that while AI tools themselves are becoming more widely available, the ability to embed them effectively remains uneven.
Access the report
Fill in your details to access our full report
Why Data Will Define the Next Phase
Across the survey findings, one theme is consistent. While confidence in AI’s potential is widespread and adoption is progressing, the ability to translate that potential into sustained performance depends on the quality, structure and integration of underlying data.
The organisations best positioned to succeed are those treating data as a strategic asset, investing in consolidation, governance and accessibility early enough to support scale. In this context, AI advantage is less a question of innovation than of execution. The next phase of value creation will be shaped by the strength of the foundations on which AI is built.