AI and investment analysis

The traditional hierarchy of investor relations has been completely upended. For decades, the single source of corporate truth was the annual report, which was followed closely by the investor presentation and the regulatory news service. It was assumed that if the copy was polished and the charts perfected, the message would land. But a fundamental shift in the investor research workflow has resulted in this assumption becoming obsolete.

 

At a recent industry briefing, a senior portfolio manager at a global investment firm shared a harsh reality. He no longer begins his analysis by opening a PDF or scrolling through a corporate website as he may have in the past, but instead, he feeds the entire bulk of information about a company into an AI bot. By the time he walks into a meeting with management, his investment thesis is already semi-created, based on a machine-generated summary that the company’s IR team never even saw, let alone briefed.

 

 

The age of digital invisibility

We have moved beyond the era of the Search Engine and into the era of the Answer Engine. In our old world, the goal was to secure the top blue link on a Google search results page. However, in the new world, the goal is to be the primary cited source in an AI-generated overview.

 

If your corporate narrative is not structured to be machine-readable, your company effectively ceases to exist in the investor’s primary discovery layer. Research suggests that as many as 4 in every 5 (80%) of searches in travel now result in an AI overview before a user ever clicks a link (Kantar, 2024), also called ‘zero click’. For a CEO or MD, the risk is clear: if the AI does not agree with your business growth strategy because it can’t find the relevant evidence in a format it understands, then in the eyes of the investor, that strategy simply does not exist.

 

 

Why generic content is a valuation risk

There is a temptation to fight fire with fire by using generative AI to churn out endless streams of corporate content. But this is a strategic mistake. While AI is excellent at processing data, it is increasingly sensitive to signals of EEAT: Experience, expertise, authoritativeness, and trustworthiness.

 

Generic, AI-generated filler actually harms stock valuation by diluting these signals. Modern algorithms and sophisticated investors are looking for agentic-ready content: information that is structured, unambiguous, and backed by verifiable evidence. When a company relies on platitudes rather than specific metrics and clear strategic pillars, the LLM (large language model) fills the vacuum with its own interpretations or, worse, data from third-party sources that you can’t control.

 

The shift to agentic optimisation means we must treat the machine as our first audience. This doesn’t mean writing for robots; it means ensuring that your single source of truth - your corporate website - is technically accessible and has a clear language structure. If you use the word ‘fragrance’ while the rest of the industry and the LLMs are looking for scent, you lose visibility (Essity/MRS, 2024). Precision in language has never been more closely woven into market perception.

 

 

Regaining control of the narrative

To regain control in this landscape, business leaders must now pivot from traditional broadcasting to more strategic signal management. This will involve three key shifts in mindset.

 

First, recognise that your corporate website is no longer just a marketing tool; it is your official record for the machines. Every FAQ, sustainability metric, and leadership biography must be formatted to be easily ingested and cited by bots. If an AI agent cannot cite you, it will cite your competitors or a potentially biased analyst report instead.

 

Second, the human in the loop is more important than ever. While we must optimise for machines, the weight of content attributed to credible, human experts, including CEOs, CFOs, and technical leads, is increasing. In a sea of machine-generated noise, the human imprint is the biggest trust signal.

 

Finally, we must move from monitoring search results to monitoring answer engine citations and visibility. We need to know not just where we rank, but how we are being summarised by AI. If the machine’s summary of your five-year plan misses the core value driver, the problem isn’t the AI; it’s the lack of clear, structured signals and information in your communications strategy.

 

The investor is already here, but this time they’ve brought a machine with them. The question for the C-suite is no longer ‘what does our annual report say?’, but instead ‘what is the machine telling our investors about us?’ The answer to that question will define the next decade of corporate valuation and success.

 

---

 

Simon Gittings is Head of IR and Corporate Communications at IDX

 

Main image courtesy of iStockPhoto.com and NanoStockk