Galata Wind’s Müge Yücel shares tips, tricks and prompts for matching the buy side’s increasing use of AI
If you remember the 1999 sci-fi classic The Matrix, humanity eventually realizes that to survive in a world governed by sentient machines, they must deploy their own autonomous programs to fight back. If you have been in investor relations long enough, you might feel as though we are waking up to a similarly stark reality in 2026.
We have officially entered the ‘agent AI vs agent AI’ paradigm.
When a portfolio manager’s first line of screening becomes fully automated, we as IROs have no choice but to deploy our own AI agents to curate the data those buy side machines are consuming. As the buy side rapidly integrates artificial intelligence into its workflows, the entire philosophy of shareholder targeting is shifting beneath our feet. This is no longer just about capturing capital; it is about ensuring we are not shouting our investment thesis into the void.
The paradigm shift: from static rolodexes to predictive engines
For decades, targeting required extensive manual effort. We relied on static lists, often updated only once a year, focused on ‘active’ funds, filtered by geography. We depended entirely on brokers for corporate access. It was a world driven by gut instincts, where our CRM served as a digital rolodex.
Today, targeting is no longer just an administrative task for organizing roadshows; it is a strategic, data-driven capital allocation process. We need dynamic targets updated quarterly. We must consider the combined influence of active, passive and quantitative capital, emphasizing thesis-led matching based on real-time engagement signals and fund flows. The shift from traditional peer analysis to predictive machine learning represents a quantum leap in what our IR teams can accomplish.
The buy side is already automated
The catalyst is the buy side’s aggressive adoption of AI. According to the 2026 Global InvestOps Report by SimCorp, more than two thirds of investment managers use AI in their front-office operations, while private equity adoption has surpassed 82 percent.
Targeting is no longer just an administrative task for organizing roadshows
The impact on IR is significant. Institutional investors are increasingly by-passing traditional sell-side research reports. Instead, their predictive machine learning algorithms process real-time global ESG scores, news sentiment and macroeconomic data to identify targets and divestment triggers – often before we are even aware a shift is occurring.
New challenges for IROs in the AI era
This technological acceleration brings a unique set of challenges for corporate issuers:
- Workflow integration gaps: Tools are ineffective without centralized data. You cannot run predictive models on fragmented calendars and offline Excel sheets. Integration across the technology stack is essential
- The ‘black box’ myth: We initially feared AI targeting was an opaque black box. That has been thoroughly debunked. Modern tools provide transparent, interpretable and rational justifications for investor matching
- The mechanics of passive capital: We can no longer ignore passive money. While index funds will not request meetings, their algorithmic, mandatory purchases significantly alter our liquidity, reduce volatility and permanently change our shareholder structures
- Navigating ‘AI-washing’: As AI becomes a board-level imperative, companies are increasing disclosures. However, regulators are monitoring closely. The SEC has made ‘AI-washing’ – overstating capabilities or rebranding basic automation – a core enforcement priority, creating significant reputational risks, according to Freshfields.
Strategies for the modern IRO: surviving the ‘agent AI’ era
To thrive today, we cannot simply do the old things faster; we must fundamentally change how we engage with the market. The new playbook requires a shift from human-to-human networking to machine-to-machine curation. Here is how to recalibrate your targeting for the AI era:
- Deploy your own ‘agent AI’ for data curation: If the buy side’s first filter is a machine, your first line of defense must be one as well. Institutional investor AI assistants are scraping the web for global ESG scores and real-time news sentiment. Modern IR teams must deploy their own AI agents to actively monitor this sentiment, analyze the tone of market reports and generate optimized summaries. You are no longer just tailoring a pitch deck for a portfolio manager; you are formatting your data to be perfectly ingested by their algorithm
- Upgrade from peer analysis to predictive machine learning: Traditional targeting relies on ‘cross-shareholder peer group analysis’ – identifying who owns your direct competitors. This approach is too slow and limits your scope. The new standard uses predictive machine learning to execute behavioral matching, allowing you to discover entirely new pools of globally aligned capital that you would never have found by simply examining your local sector rivals
- Shift from static lists to real-time triggers: The days of downloading a quarterly targeting list from your broker are over. The buy side’s algorithms are identifying targets and divestment triggers in real time. Your targeting apparatus must be equally dynamic, relying on real-time data processing to track algorithmic and index-tracking fund flows as they occur
- Optimize your disclosures for the algorithm: Generative AI rejects ambiguity. When it scrapes your filings and sustainability reports, it seeks hard data. Eliminate generic sustainability language. Reformat your ESG narrative to focus strictly on financial utility: capital expenditure discipline, carbon transition plans and risk management. If your disclosures are not quantifiable, the buy side’s machine learning models will simply skip over them.
The IRO’s tactical AI toolkit: four quick-fixes to implement
Understanding the ‘agent AI’ paradigm is one thing; operationalizing it is another. For IROs ready to modernize their targeting workflow, here are four specific, immediate ways to use AI to your advantage:
- The ‘reverse-engineering’ notebook strategy: Stop guessing what a target fund cares about. Gather the target fund’s last three letters to shareholders, their recent proxy voting guidelines and your company’s latest earnings transcript. Upload all of these into a secure, closed-loop AI tool
- Prompt the AI: ‘Act as a critical portfolio manager at [Fund Name]. Based on your recent shareholder letters, identify the top three gaps in our current earnings narrative. What specific financial or ESG data points would convince you to initiate a position?’ This transforms a static fund profile into a dynamic, personalized targeting strategy
- Optimize your disclosures for ‘machine readability’: The algorithms scraping your press releases do not care about flowery adjectives; they are searching for hard data
Before publishing your next quarterly update, run the draft through an LLM
Before publishing your next quarterly update, run the draft through an LLM and use the prompt: ‘Extract all quantitative financial and sustainability metrics from this text into a standardized table’. If the AI struggles to format the data, the buy side’s algorithms will as well. Structure your releases with clear headings and bulleted KPIs
- ‘Red-team’ your roadshow pitch: Before putting your CEO or CFO in front of a tier-one target, use AI to pressure-test the meeting. Feed an AI the transcripts of the top three bearish analysts covering your sector, along with the target fund’s historical investment criteria. Instruct the AI to act as an aggressive buyside analyst and generate the five most challenging, data-driven questions your management team is likely to face
- Automate ‘peer alert’ sentiment scraping: Instead of manually reading your competitors’ earnings transcripts, automate the extraction. Set up an AI workflow that ingests the quarterly transcripts of your closest peers. Prompt the AI to output a concise summary: ‘What new institutional investors asked questions on this call, and what was the underlying sentiment regarding [insert your specific macro theme]?’. This provides real-time, highly targeted lead generation.
Reclaiming strategic independence
Shareholder targeting is not about chasing marquee funds or booking an exhaustive list of meetings just to appear busy. The ultimate goal is to build a high-quality capital profile that aligns with your company’s liquidity, narrative and long-term goals.
By adopting predictive capabilities, mastering the mechanics of passive flows and engaging the automated buy side with financially quantified narratives, we can reclaim our strategic independence. Let AI handle the heavy data lifting, freeing us to focus on what we do best: the nuanced, strategic art of human relationship building.
Müge Yücel is director of investor relations & sustainability, corporate governance committee member at Galata Wind. She also sits on the IR Impact editorial board.
