Believe the Hype: How AI is Adding Value in Fund Administration

Authored by: Raj Gidvani, Chief Technology Officer
Updated on: December 10, 2024

It seems any conversation on technology today inevitably leads to AI, and how it’s going to completely transform society. The lofty promises of the technology extend to fund administration as well, but it’s critical for private equity funds to separate the hype from reality to plan effectively and reap the benefits that AI actually delivers.

Make no mistake, the ability of AI to deliver value by automating manual processes and digitizing human intuition at scale has already been proven. In our own work, we have seen AI augment human capital output by at least 25-30%. This enables faster and more accurate decision making by breaking down large volumes of data into tactical summaries and actionable insights.

While we believe that the overall reception to AI across the industry is cautiously positive, it’s important to be realistic about where and how it’s creating value, and which challenges still need to be navigated.

Where is AI Creating Value?

There are several areas where AI is delivering value in fund administration today, but they broadly fall into three categories.

- Improved efficiency and accuracy: By using AI to automate and streamline repetitive and manual tasks, such as data extraction, validation, and reconciliation, we can reduce human errors while also improving the speed and quality of our work.

For example, AI can automatically tag and categorize files based on their content or metadata and apply the appropriate access permissions and encryption levels. AI can also help find relevant files faster by using natural language queries and semantic search.

- Enhanced insight and value: By using AI to analyze and synthesize large and complex data sets, such as financial statements, market data, and industry trends, we can generate deeper and more meaningful insights and recommendations for clients to provide more value-added services and solutions that address their specific needs and challenges.

For example, AI can automate knowledge sharing by generating summaries, previews, and recommendations based on the recipient's profile and preferences as well as data visualization and dashboarding solutions from the data.

- Increased innovation and differentiation: By using AI to create and leverage new capabilities and opportunities, such as natural language generation, we can improve products and processes.

For example, an AI chatbot can be trained on all of a company’s various policies and then offer specific answers when queried.

AI can also add value for client data protection, particularly at the end user’s desktop, by detecting, preventing, and proactively responding to potential breaches.

Quantifying the AI Impact

We conducted proprietary, controlled tests of AI’s efficiency and impact on fund administration that showed significant improvement. We have quantified this value add by measuring the improvement in key performance indicators (KPIs) such as:

• The time and cost savings achieved by reducing the manual workload and increasing the automation level of data extraction and structuring tasks.
• The increase in data accuracy and completeness by reducing the error rate and increasing the coverage and confidence of the extracted data.
• The increase in customer satisfaction and retention by delivering high-quality reports and data in a timely manner and providing better service and support.

If you can cut 20 to 25% of the time it takes in the review process for every statement and related document over the course of the year, that not only adds up to some significant savings, but that time can be applied to higher value work. As you deal with more entities, catch more errors, and consider additional documents, the potential time savings scale.

Challenges Remain

While the value of AI is evident, we have a long way to go and several immediate challenges to solve on the way to wider adoption and deployment. First among these is data quality and consistency: the PE space is known for complicated deal structures in funds, including their legal entity hierarchies, side letters, etc. There is less standardization in PE processing and administration, which creates challenges for AI.

For example, different funds may use different terms or definitions for the same concepts, such as performance metrics, fee calculations, or investor types. This can make it difficult to apply AI models or algorithms that rely on common data schemas or labels. These can be tackled through a data harmonization process that maps the data from various sources and formats to a unified and consistent representation, using a combination of rule-based and machine learning techniques. This can allow a large and diverse data set to be leveraged for AI applications, while preserving the accuracy and integrity of the original data.

Data security is always a consideration when applying AI, especially in the PE space, where sensitive and confidential information is involved. Combining global client data also has GDPR implications, as well as other data protection regulations in different jurisdictions. Ensuring this data is protected from unauthorized access or misuse is paramount.

Fund administrators must implement a comprehensive data security framework that follows the industry’s best practices and institutional standards, such as encryption, authentication, authorization, auditing, and backup.

In addition to the technical and operational challenges, applying AI in the PE space also raises ethical and social challenges, such as fairness, transparency, and accountability. Firms must have a responsible AI policy that sets the guidelines and principles for the design, development, and evaluation of AI solutions.

The policy should cover topics such as data quality, security, and privacy, as well as algorithmic bias, human oversight, and stakeholder engagement. Defining the roles and responsibilities of the different teams and individuals involved in AI projects, as well as the mechanisms and processes for monitoring and reporting AI performance and impact will be critical to success.

We think that AI will continue to evolve and improve and allow for more advanced and customized services and solutions, such as predictive analytics, natural language processing – where computers understand and generate written and spoken language, and computer vision – where computers look at text, images, and videos, and extract information. These will help to enhance client experience and satisfaction, by providing more personalized, timely, and relevant information and insights, and by creating more interactive and engaging communication channels.

By identifying and mitigating risks and fostering innovation and growth, AI can help us to create more value and impact for our clients and society, but managers should still be aware of what is hype, and what is reality.

Originally published in PEI Private Funds CFO

Published on: December 3, 2024