Moving Past Roadblocks To Unlock Artificial Intelligence in Financial Services

Addressing Persistent Tensions Surrounding AI Use in the Industry

  • November 27, 2025

There is a persistent tension surrounding artificial intelligence use in the financial services industry. An Economist Impact report found that while four in five executives in the industry agree that AI use will distinguish winners from losers, nearly two-thirds also believe that the complexity and risks often outweigh the benefits. Indeed, because financial institutions are heavily regulated, their risk for how data is used and how technologies are applied to customer-facing services is higher than what organisations in other industries may encounter.1 Safe AI innovation is not out of reach for financial services, though, and there are several use cases for which these organisations can tap into new technologies without exacerbating risk.

Many practical applications for AI in financial services are simply the next evolution of existing uses of AI. Initial use cases that can be implemented in contained, low risk ways, while providing significant impact, include the following.

  • Disputes and investigations. In discovery and document review, legal teams are increasingly testing the efficacy of various generative AI tools. Using a clean and safe sample of data, organisations can test large language models without unnecessary risk exposure or ethical concerns. One approach is to apply generative AI to review relevant documents from a previous matter and compare the results against the human review decisions. This provides a baseline for how effectively the AI performed against existing methodologies. Generative AI tools are also capable of pre-identifying privilege based on parameters set by the legal team to help automate logging of privileged material.
  • Repapering exercises. Optical character recognition capabilities can significantly speed up the process of analysing handwritten materials and low resolution scans in a variety of matters, including cumbersome repapering exercises. By automating the most manual steps of contract review or repapering, particularly those where generative AI has been demonstrated to have a high degree of reliability, legal teams can reduce the time and cost required to complete the task without sacrificing quality and accuracy.
  • Compliance. Generative AI models can be designed to support summarisation and reporting across structured databases for compliance purposes. For example, models can ingest large volumes of structured data, such as a set of transactions or financial accounts, and summarise and report on trends and anomalies in the data. When compliance teams are rigorous in ensuring correct, quality data is fed into the model, these tools can improve speed to insight without adversely impacting sensitive information. Additionally, generative AI tools can automate speech to text cleanup and review by removing errors and filling in gaps from transcriptions of recordings, making them easier to review and analyse for compliance purposes.
  • Data subject access requests. Large language models are showing significant results in their ability to identify and extract personal information from large data sets. For financial organisations that must comply with various global privacy laws, this can make it much easier to find and produce information containing specific personally identifiable information, reducing the time and cost of responding to data subject access requests.

Given that financial institutions often face a much higher degree of risk than organisations in other industries, it’s critical to balance risk management with AI innovation. Global AI regulation is also evolving, with the EU AI Act serving as a benchmark in the kinds of requirements authorities are likely to enforce around the use of AI.2 Explainability is one key area and legal teams should proactively review existing and impending uses of AI to ensure they meet explainability standards and that the models in place are not violating privacy requirements, financial reporting rules and other statutes.

Legal and technical experts can partner to establish parameters for how AI may be used in legal and compliance settings and to guide governance around new technology deployments in other parts of the business. Bringing technical and legal minds together will be critical in supporting financial services institutions in achieving their AI objectives without introducing excess risks.

Footnotes:

1: Kingsley, Jeremy. “Banking on a game-changer: AI in financial services,” Economist Impact, (March 16, 2022).

2: The EU Artificial Intelligence Act.

Related Insights

Related Information

Published

November 27, 2025

temp Key Contacts

Jon Chan

Senior Managing Director

Most Popular Insights

  1. Navigating Tariffs: 10 Strategies to Protect the Bottom Line
  2. The Hidden Risk for Data Centers That No One is Talking About
  3. It’s La La Land Again in Financial Markets
  4. Geopolitical Risk Management: No Longer Just a “Nice to Have”, But a Corporate Imperative
  5. Corporate Sustainability Report

Sign up to get access to FTI Consulting Insights

Subscribe