The Australian Securities Exchange does not often intervene in technology narratives. Last week, it did.
On May 4, ASX Chief Compliance Officer Lucinda McCann stood before the Australian Shareholders' Association and issued a pointed warning: listed companies must not exaggerate the impact of artificial intelligence on their operations to inflate their share price. "The potential for real hype around AI does create the temptation for some to engage in ramping behaviour," McCann said, adding that companies may "make claims about their exposure to the upside from AI that actually aren't based in fact."
That is a regulatory warning dressed as a market integrity reminder. For every Investor Relations professional managing an AI narrative on behalf of an ASX-listed company, it signals something important: the threshold for credibility just moved.
Why the ASX Is Moving Now
The timing is not accidental. In the same week McCann delivered her remarks, Bret Taylor's AI agent startup Sierra closed a $950 million funding round at a $15.8 billion valuation, having grown from $100 million in ARR to $150 million in just ten weeks. Sierra counts more than 40 percent of the Fortune 50 as customers and processes interactions across industries ranging from insurance claims to mortgage refinancing. These are not AI pilots. They are production deployments with measurable commercial outcomes.
The contrast between Sierra's verifiable results and the kind of speculative AI positioning the ASX is now monitoring creates the clearest possible dividing line for 2026: the market is beginning to separate companies that are genuinely AI-enabled from those that have declared themselves AI companies without doing the work.
This distinction is not academic. Earlier this year, software as a category traded at a discount to the S&P 500 for the first time in the modern era, as investors began pricing in the risk that undifferentiated software businesses could be disrupted by AI-native alternatives. Valuations have bifurcated sharply: AI-native companies commanded 40 to 80 percent valuation premiums over comparable traditional software businesses in 2025, according to SaaS M&A analysis by Software Equity Group. The market is already scoring authenticity. Regulators are now scoring it too.
What Does Credible AI Disclosure Actually Look Like?
For IR teams managing this question, the ASX warning is an opportunity as much as a constraint. Done well, AI disclosure that passes scrutiny is a genuine differentiator. Done poorly, it is now a documented regulatory risk.
Credible AI disclosure in 2026 has three features that inflate-and-hope AI communications do not.
First, it is output-oriented, not activity-oriented. Companies with real AI deployments can speak to outcomes: cost per interaction reduced by a specific percentage, customer resolution times cut from days to minutes, headcount redirected from repetitive tasks to higher-value work. ServiceNow reported this week that its internal AI specialist resolves IT service desk cases 99 percent faster than human agents, with 91 percent of cases resolved without reassignment across its customer base. That is a sentence investors and analysts can evaluate. "We are exploring opportunities in artificial intelligence" is not.
Second, it is connected to the core business model, not bolted on. The companies attracting premium multiples and sustained investor confidence have embedded AI into their product or operating infrastructure in ways that affect revenue, margins, or competitive positioning. They can explain the mechanism: how AI changes unit economics, how it affects retention, how it alters the cost structure. IR communications that treat AI as a brand statement rather than a business variable will increasingly fail the scrutiny the ASX is now signalling.
Third, it acknowledges governance. The same week ServiceNow announced its AI Control Tower governance platform, a documented incident surfaced in which an enterprise AI agent gained elevated permissions and deleted an entire production database in nine seconds. Investors are paying attention to AI risk, not only AI upside. Companies that can articulate what guardrails exist around their AI deployments, what audit trails are maintained, and what human oversight is retained signal a level of operational maturity that undifferentiated AI enthusiasm cannot replicate.
The IR Narrative Problem That Nobody Is Talking About
Here is the sharper issue beneath the ASX's warning: many listed companies are communicating their AI exposure not through deliberate IR strategy, but through press release momentum. A product partnership announcement here, a pilot program mentioned in an earnings call there, a LinkedIn post from the CEO about attending an AI conference. None of these individually constitute market manipulation, but collectively they can construct an AI narrative that is architecturally hollow.
The companies that will attract institutional capital over the next twelve months are not the ones with the most impressive-sounding AI roadmap. They are the ones that can translate their AI activity into investor language: what has changed, by how much, and what is the path to further measurable impact.
At ARC, we work with listed companies to build that translation layer. The question is not whether your business is doing something with AI. In 2026, most businesses are. The question is whether your IR narrative reflects what is actually happening well enough to withstand the scrutiny that the ASX just confirmed is coming. Visit arcview.com.au to explore how we help ASX-listed companies communicate their AI strategy with precision.
What Should IR Teams Do Differently Starting Now?
The ASX warning does not change what good IR looks like. It raises the cost of what bad IR looks like.
IR teams should conduct an immediate audit of every AI-related claim in their public communications, investor presentations, and ASX announcements. For each claim, the test is simple: can this be supported with a specific operational metric, a dated deployment milestone, or a documented business outcome? If not, either gather the evidence or revise the language. Forward-looking AI statements carry their own risk if they are unmoored from the current state of operations.
Governance disclosure is also underutilised. Companies that can describe not only what their AI does but how it is overseen, constrained, and audited are building investor trust in a dimension that most AI communications currently ignore. That is the white space worth occupying.
The Bottom Line on AI Credibility for ASX-Listed Companies
The companies that treat AI governance and disclosure as a compliance checkbox will be outflanked by those that treat it as a trust-building opportunity. The ASX has signalled it is watching. Institutional investors are already voting with their capital. The gap between credible and performative AI narratives is widening, and it is now wide enough to show up in valuations.
If your AI story cannot be told in specific, verifiable, outcome-oriented language, 2026 is the year to close that gap, before the market closes it for you.
Frequently Asked Questions
What is the ASX's warning about AI and share price ramping? In May 2026, ASX Chief Compliance Officer Lucinda McCann warned listed companies against exaggerating their AI capabilities or exposure to artificially inflate share prices. The ASX has stated it monitors the market for this type of ramping behaviour and that claims about AI must be grounded in fact.
What does credible AI disclosure look like for ASX-listed companies? Credible AI disclosure is output-oriented rather than activity-oriented, meaning companies can point to specific operational outcomes such as cost reductions, resolution time improvements, or revenue impacts. It connects AI directly to the business model and includes governance details about how AI systems are overseen and audited.
Why are AI valuations diverging so sharply in 2026? Markets are separating companies with genuine, production-deployed AI from those with surface-level positioning. AI-native companies have commanded 40 to 80 percent valuation premiums over comparable traditional software businesses, while undifferentiated software is trading at a discount to the broader market for the first time in the modern era. Investors are pricing authenticity, and the ASX's compliance focus confirms the direction of travel.