Over the next year, 37% of asset management companies will focus on expanding artificial intelligence (AI) use, 22% will increase experimentation, and 28% will monitor progress, according to Linedata.
“In a market under multiple pressures, AI adoption in asset management is accelerating,” Jamil Jiva, Global Head of Asset Management at Linedata, said. “More companies are investing in this technology to stay competitive and develop new use cases,” he added.
Current State of AI in Asset Management
According to Linedata’s latest report, “What’s Next: A closer look at Artificial Intelligence in Asset Management,” currently, 32% of asset management firms have not yet started their AI journey, while 33% are in the experimentation phase, and 36% are actively using AI.
Among the latter, 14% have multiple use cases in production and plan further implementations.
Popular Applications of Generative AI
The most popular applications of Generative AI (GenAI) include document synthesis (28%), data extraction (28%), and knowledge bases/Q&A (17%). Additionally, Generative AI is finding crucial commercial applications, particularly in enhancing front office productivity.
Optimizing the efficiency of the transaction team (23%) and directly generating transaction research or yields (18%) are among the most common use cases. AI also plays a vital role in boosting the productivity of middle and back-office operations (19%), the report found.
Challenges in AI Implementation
“Implementing AI solutions and realizing their value is a technological endeavor that requires time to establish the right foundations, secure buy-in, drive cultural change, and mitigate risks,” commented Jiya.
The survey also found that for 46% of respondents, AI expertise comes entirely from within the company, while only 14% rely solely on external partners. The remaining 40% use a hybrid approach.
Accessing AI Solutions
AI solutions are primarily accessed by purchasing off-the-shelf products (25%), though a notable 18% develop them entirely in-house. Conversely, 32% access AI solutions indirectly through brokers, fund administrators, or outsourcing service providers.
Data Quality and Challenges
Besides AI adoption overall enthusiasm, challenges are numerous: data quality and update (19%), costs understanding and business cases development (15%) along with AI expertise availability (13%). These challenges also do appear when it comes to AI solutions extension.
“Most companies prefer to tightly control their AI capabilities. However, developing in-house expertise is challenging, and no single solution can address all a company’s needs. This has led to the rise of hybrid approaches that combine internal resources with external partnerships,” Jiya said.
He added that data quality clearly appears as a major challenge to address. “Data must be reliable, consistent, safe, and easily accessible to be used efficiently training Large Language Models (LLMs), which clearly sits as a massive project in light of heterogeneous systems that spans across the financial institutions,” he said.
“Many asset managers are now developing data lakes, which proves a complex project that requires clear objectives,” Jiya added.
Survey Insights
The survey involved nearly 100 fund and asset managers, primarily focusing on hedge fund and private equity investment strategies. Data was collected via an online questionnaire and in-depth interviews during the first quarter of 2024.
Respondents held various executive positions in companies of different sizes and geographic locations: 40% from North America, 40% from Europe, 15% from APAC, and 5% from other regions.
- 0 Comments
- Ai Process
- Artificial Intelligence