As artificial intelligence, machine learning, and deep neural networks integrate into institutional investing, their true potential sparks active debates. Are they merely enhancing operational efficiency, or are they revolutionizing investing models and strategic portfolio decisions? Below are some of the compelling questions our LP clients are exploring.
- What is Vidrio’s view on artificial intelligence and how will it impact the allocator space?
- What level of complexity do institutional investors wish to take on in rolling out an AI plan, and why is looking at it simply as an IT expense the incorrect approach?
- How can we address cybersecurity concerns at the organizational level, and is there a way to avoid overreliance on AI algorithms?
High-level perspective on AI for allocators
AI can transform the investment industry by enabling faster data collection and improved decision-making. Allocators have access to abundant data, but it's often unstructured. Proper AI implementation helps analyze this data efficiently, benefiting even those without tech expertise.
Second, AI through natural language processing will assist allocators in asking questions and enabling decision-making through corresponding answers. In our view, it will eventually evolve into something more mainstream like Siri or Alexa. Instead of asking about the weather or what new show is on, it will respond to requests for assistance in uncovering new investment opportunities. Risk management should benefit from this evolution, as CIOs and portfolio teams can inquire about gaps in the portfolio on a trending alternative asset class, style drift from the initial fund or portfolio mandate, or even how historical scenarios might influence future portfolio performance.
AI complexity dilemma
AI plan execution for those in the asset allocation space should be understood based on how much complexity your investment organization is willing to take on compared to your current operations and goals. Don’t view it as an update or ongoing maintenance of Outlook or your notebook computers; it should be much larger than that and more strategic.
In today's fast-paced business world, allocators must understand the buy vs. build debate. The key challenge is integrating AI smoothly across the organization, not just within the IT department. Organizations that take this narrow approach might see short-term gains, but they risk missing out on broader opportunities. By involving a diverse range of team members from the outset, companies can address AI development more effectively and tackle significant investment challenges, ultimately, fostering a culture of innovation, continuously enriched with AI knowledge and training. We believe this is the key to unlocking transformative growth.
When asked about the buy vs. build debate in our 2024 report, respondents were evenly split on the direction to take, while a smaller group preferred to push AI execution plans down the road, continuing with legacy tools. Reading into these results we believe that a paradigm shift is necessary, and investors should not look at AI as a simple tool but as an integrated platform that informs multiple parts of your business across department silos. I elaborate more on these ideas in the Improving Alpha podcast with host Michael Oliver Weinberg and encourage Pensions & Investments readers to listen to that episode.
Cybersecurity and overreliance on AI
In Vidrio’s recent survey report of allocators’ opinions on AI, we explored whether allocators have incorporated AI-based tools or technologies into their investment strategies. Overall, 53% of the respondents said that they were actively harnessing AI to manage areas of investment operations. Yet, challenges remain in AI privacy concerns and an overreliance on this type of technology.
Robust cybersecurity is essential due to the rapid innovation of AI models. Key concepts are needed in a plan to protect allocator data from threats. These include regular cadence of testing within industry standards and baselines; scaling testing procedures to understand the latest threats and real-world attack scenarios and determine who inside the organization is responsible for testing; and enhancing organizational awareness on how cyberattacks are becoming more complex with smarter malware.
The second point on AI overreliance is a valid concern among allocators, yet we believe it can be adequately managed through the combination of human intervention and AI machine learning. For example, consider an AI engine where algorithms have been trained on a given format of performance statements, monthly manager letters, Institutional Limited Partners Association reports, and capital calls and distribution letters. The engine and algorithms will constantly look for known fields to enhance and enrich portfolio intelligence. When a new statement enters the AI engine that does not follow the training data, a flag will be raised for human intervention. Without that alert, the AI engine would fail on its own and processing would grind to a halt. Through human intervention, the AI engine can be recalibrated, further enhancing its learning capabilities and avoiding a similar disruption in the future. This synergy between human expertise and AI is at the core of Vidrio’s approach. Taken a step further, as these internal data sets expand, enhanced by human intervention, the new data set will become more of a commodity for allocators embracing AI. Essentially, creating a secured single source of truth based on real-world first-party performance data that operates outside of more public AI engines like Chat GPT, Claude and others.
Ultimately, the decision and level of involvement of AI in allocator investment operations will require research and understanding from stakeholders across all business levels. There is not a one-size-fits-all approach in obtaining an optimal total portfolio view. Execution is critical and transparency into artificial intelligence execution plans will go a long way as regulations address future innovations.
Mazen Jabban is founder, chairman and CEO of Vidrio Financial. He also is an adjunct professor of finance at NYU’s Stern School of Business, lecturing on alternative investments. He is based in New York. This content represents the views of the author. It was submitted and edited under Pensions & Investments guidelines but is not a product of P&I’s editorial team.
