The Revenue Imperative: Why AI Must Move from Efficiency to Efficacy

Intelligent Origination

Private credit is expanding at a pace that is restructuring the competitive landscape of institutional lending. With global AUM now estimated at US$3.5 trillion and still growing, the market is drawing capital away from traditional bank intermediaries. As a result, the expectation is that lenders can move faster, originate smarter and deploy more precisely than ever before. Against this backdrop, artificial intelligence has become the defining technological story in financial services.

Global AI spending is forecast to reach approximately US$2.5 trillion in 2026, with banks and asset managers among the highest-spending sectors. A meaningful gap has opened up between investment and impact. For most institutions, AI remains a back-office productivity story — a tool for cost reduction rather than a driver of competitive advantage. That is beginning to change and the institutions that recognise the shift earliest will be the ones that define the next decade of private credit.

From Efficiency to Efficacy

The current wave of AI deployment in financial services is largely defensive. Cost management and process efficiency dominate current adoption. A survey by the Bank of England of UK financial institutions found "optimisation of internal processes" to be the main use case for AI.

In North America, the focus on efficiency is even more pronounced. In its Summer 2025 edition, Coller Capital's Global Private Capital Barometer found that 90% of investors highlighted "streamlining internal processes, enhancing productivity and optimising resource allocation" as the primary benefits of AI. Automation is real, but it is not transformation.

The more consequential shift is now underway. At JPMorganChase, business-side leaders are being elevated to run the AI mandate, not with a remit to cut costs, but to rethink how the institution generates revenue. As one of the firm's CIB co-CEOs put it plainly at a recent industry conference: the excitement around AI has moved decisively to the revenue side.

Around 70% of financial services executives share this view, expecting AI to directly contribute to revenue growth in the coming years. Yet most AI deployments remain anchored to basic process automation. The implication is straightforward: the institutions willing to move AI into front-office origination will pull ahead of those that do not.

Chart note: 41% of respondents are using AI for optimisation of internal processes. (Bank of England, Artificial intelligence in UK financial services, 2024)

The Structural Challenge of Origination

Private markets have long operated in an environment of informational complexity. Unlike public markets, which run on structured, standardised data flows, private credit originators must synthesise insight from emails, legal documents, unaudited financials, sector commentary and management call notes, few of which speak naturally to one another.

The core origination bottleneck is due to a shortage of information. It is the cost, in analyst time and cognitive bandwidth, of organising dispersed data into a coherent credit view. Large language models resolve this by acting as a transformation layer, ingesting and structuring disparate inputs into a queryable, comparable dataset that removes the friction without removing the professional judgment.

The speed differential is striking. AI systems can identify relevant origination targets almost two orders of magnitude faster than a human analyst can evaluate a single prospect. The result is not a replacement of the analyst, it is a dramatically better starting point, freeing expertise for the decisions that require it: qualitative assessment, relationship management and deal conviction.

The Window is Closing

At least 80% of private equity workflows now incorporate technology in deal sourcing and 95% of these firms plan to increase their AI investment materially over the next 18 months. Yet just 7% of institutions have achieved full AI integration across their operations. The gap between aspiration and execution is real — and so is the competitive advantage available to those who close it first.

The institutions now embedding AI into origination are not running experiments. They are reshaping origination: how deals are sourced, how prospects are screened and how conviction is built. The research phase is ending. Execution has begun.

References

1 — The Revenue Imperative: Why AI Must Move from Efficiency to Efficacy

• [1] AIMA, Strong growth sees private credit market reach US$3.5 trillion (2025)
• [2] Gartner (2026)
• [3] Bank of England, Artificial intelligence in UK financial services (2024)
• [4] Coller Capital (2025)
• [5] JPMorgan Chase & Co., Presents at UBS Financial Services Conference 2026 Transcript (2026)
• [6] WEF, Artificial Intelligence in Financial Services (2025)
• [7] WEF (2025)
• [8] WEF (2025)
• [9] WEF (2025)

‍