Incorporating AI into a Public Equity Manager’s Investment Process


Joseph Byrum

May 10, 2024

Share on:

It is widely accepted that AI will be transformative to the investment industry, creating both disruption and opportunities. The changes will permeate investment managers and the organizations they interact with, from clients such as pension plans, family offices, endowments & foundations, to consultants and data providers. AI will raise the bar for all fiduciaries, facilitating access to greater transparency while also enhancing analytical tools. Change can be difficult, and for investment managers it can even be frowned upon as investors and consultants prefer consistent processes from which to evaluate managers. Nevertheless, investment managers will need to incorporate AI applications at a minimum to enhance productivity, manage costs, and hopefully to support decision making and performance.

Just as the spreadsheet revolutionized business analytics in the 1970s, enabling managers to quickly test scenarios and find answers to complex questions, AI is poised to transform the field of investment management. It will find its way into all corners of the investment organization, with applications ranging from automating back-office processes to complicated algorithms designed to generate new sources of alpha. Determining how best to utilize AI will evolve as the speed of advancements is rapid; there will be some trial and error as the technology improves and firms learn from their missteps. In this paper, we touch on elements that are common to most firms’ investment processes and highlight how Consilience’s AI-driven quantitative linguistic platform, AlphaIQ, can be used to support and enhance the investment capabilities for public equity managers.

Screening Investment Universes and Generating Ideas

Consilience’s Linguistic Risk Index identifies companies with improving, stable, or deteriorating risk profiles. This tool can be used to help managers screen out companies for consideration or identify companies for inclusion in a portfolio. The tool is particularly adept at flagging companies poised to underperform the market based on material movement in their risk score.

Security Selection Validation

Portfolio managers, analysts, and investment processes all have biases. AlphaIQ can play a role as a neutral party within the context of investment decision making. It can serve as a validation or sanity check to assess whether the investment thesis for a stock holds up to another layer of scrutiny. By leveraging advanced algorithms to integrate vast troves of structured and unstructured data, AlphaIQ can analyze companies with a comprehensiveness and impartiality that dwarfs any human analyst’s capabilities. Transcending conflicts of interest, it dispassionately assesses fundamental long-term value propositions rather than succumbing to behavioral biases around overweighting recent events or being swayed by charismatic leadership.

Supporting Alpha Signals

The linguistic model has 45 distinct alpha signals on items related to corporate event risk, bankruptcy risk, cash flow metrics, behavioral signals, and more. Managers can use the signals that align with their investment process to confirm a stock’s attractiveness or guide them to better alternatives. The signals open new opportunities for managers to move beyond the commonly used quality, growth, momentum, and value factors to a new set of signals uncovered through quantitative linguistics.

Accelerating Research

AlphaIQ provides updated generative AI reports on 3,000 domestic stocks a few days after the release of any new regulatory document. These reports highlight topics of increasing and decreasing importance to a company’s management team, along with updated company risk scores. These scores and reports can help portfolio managers prioritize a research team’s efforts, identifying risks and opportunities faster than what is physically possible for a team of research analysts. No human could possibly read everything, but a machine can.

Company Research Reports

For most fundamental equity managers, the end product that results in a buy decision is a detailed profile or company report. AlphaIQ generates executive summary (4-6 pages) and detailed company profiles (10-15 pages) on every company in the universe. These reports can supplement existing reports or be used as a reference to guide the analyst’s write-up.

Support Equity Analysts

Equity analysts must read, or by necessity skim due to time constraints, hundreds or thousands of pages of regulatory reports each quarter to keep pace with the ongoing events of the companies they follow. AlphaIQ’s generative IQ capabilities include concise responses to 18 strategic questions analysts often ask, and the responses include links to the citations within the regulatory documents where the insights were generated. AlphaIQ points analysts to where they need to be focusing their time, freeing the human mind from tedium. The system sorts through the raw data, presenting only the facts that are most relevant to the task at hand.

Risk Monitoring and Portfolio Construction

Managers can use the existing Linguistic Risk Score or create their own metric to create dashboards to monitor portfolios for changes in risk. It is common for portfolio managers to build a position in a company over time; the decision to purchase more of a security also involves a decision to sell a security. AlphaIQ can be used to confirm the merits of adding to a security and help assess which security to trim.

The Path Forward

There are many potential applications for AlphaIQ. Incorporating these tools should be part of a methodical process where they are first used to validate existing processes, then migrate to supporting and enhancing existing processes. When sufficient testing has been completed, new components and approaches can be incorporated. A measured approach will likely also play well with consultants and clients who recognize the potential benefits of AI but will need to be educated by investment managers along their journey.

Related Articles

The Rise of Small Language Models: Efficiency Meets Specialization

In the rapidly evolving landscape of artificial intelligence (AI), language models have emerged as a powerful tool for understanding and generating human-like text. While large language models (LLMs) have garnered significant attention for their impressive capabilities, a new contender has entered the arena: small language models (SLMs). These compact yet powerful models are challenging the notion that bigger is always better, offering a more efficient and specialized approach to natural language processing (NLP) tasks.


Joseph Byrum

May 31, 2024

The Perils of Complacency: Lessons from the Fallen Tech Titans

For investment managers, the message is clear: adapt or risk being left behind. The firms that will emerge as the new leaders in the industry will be those that are fast, cheaper, and better. They will harness the power of AI and other emerging technologies to streamline processes, enhance decision-making, and deliver superior results for their clients.


Joseph Byrum

May 28, 2024

Organizational Understanding of AI: Bridging the Gap for Effective Implementation

To fully capitalize on the transformative potential of AI, organizations must approach its implementation not as a singular, monolithic endeavor, but as a parallel-path operator — a multifaceted, iterative process that weaves the technology into the very fabric of the business.


Joseph Byrum

May 22, 2024