More communication is not the same as being understood.

Generative tools made producing communications cheap. They did nothing for being understood. What closes that gap is decided before anything is written.

Michael Reaney  ·  4 September 2026  ·  4 min read

Axios HQ's 2026 State of Internal Communication, its annual study, draws on responses from more than 1,200 US executives and employees and measures the gap between the organisations it classes as high performers and everyone else. What it chooses to measure is the interesting part: AI adoption, communication quality, organisational alignment, and the time lost when communication breaks down. Those four are one problem seen from four angles. Producing more has become easy. Being understood has not, and the difference is paid for in work that has to be done twice.

The pattern is relatively simple to explain and a good deal harder to fix. Generative tools have made producing communications cheap, so organisations produce more of it: more updates, more summaries, more leadership messages restated in new formats. None of that inherently becomes clearer to the intended audience. Volume and clarity are not the same output, and treating the first as a proxy for the second is where the gap opens. LexisNexis puts a number on how that pressure is absorbed: close to 60% of PR and communications respondents say they use generative tools without organisational approval, because of deadlines rather than strategy.

Our own production work points to what closes it, because the same pressure shows up daily. A leadership audio brief published every trading day, or a weekly executive interview cut from ninety minutes of raw footage, cannot rely on a fresh judgement call each time about tone, structure or what gets left out. What holds it together is decided in advance: a documented voice, a fixed cadence, and an editorial gate that checks the finished piece against both, before it goes out. That is not a quality-control step bolted onto the end of the process. It is what allows high frequency and clarity to coexist, rather than trading one for the other as volume rises.

The Axios framing points the same way from the buyer's side. It does not sort organisations by how much they publish. It sorts them by performance, and it puts alignment and lost time on the same page as communication quality. Output volume is not the variable being tested. What the function is accountable for is.

Three questions worth running against your own leadership content before the AI-driven volume increase outruns it further:

1. If output from your communications function doubled tomorrow, would internal and external understanding of your priorities double with it, or would it just get louder?
2. Is there a documented voice and cadence your team drafts against, or does each piece depend on whoever is writing it that week?
3. Does anyone check a piece against that standard before it ships, or does volume alone stand in for a quality gate?

Worth reading

• Axios HQ, 2026 State of Internal Communication report (annual, responses from more than 1,200 US executives and employees): the anchor study above, examining the gap between high- and low-performing organisations across AI adoption, communication quality, organisational alignment, and the time lost when communication breaks down. The full findings sit behind registration, so what is quoted here is the scope and framing Axios HQ publishes openly, not figures from the report itself. Axios HQ, 2026 State of Internal Communication
• LexisNexis, "AI in PR & Communications: 2026 Industry Report" (part of the LexisNexis Future of Work 2026 series, its annual study of how professionals across industries are adopting generative AI): close to 60% of PR and communications respondents say they use generative AI tools without organisational approval, driven by deadline pressure rather than strategy. LexisNexis calls this shadow AI adoption. It matters to any reader relying on a written AI policy as proof of control: the policy and the practice are visibly not the same thing. LexisNexis, AI in PR & Communications 2026
• Profound's $96m Series C (announced February, reported valuation near $1bn, 700+ enterprise customers) and Peec AI's $29m raised to date (Berlin-based, a $21m Series A on top of earlier seed funding, annual recurring revenue above $10m): the clearest recent evidence that AI-visibility monitoring, tracking how brands and executives appear in AI-generated answers, has moved from a niche measurement idea to a funded enterprise category in its own right. Worth noting for any reader evaluating these tools: they show you the picture. None of them write the thought leadership that changes it. Everything PR, on Profound's Series C

The moves

• Stellantis has appointed Stephanie Hartgrove Chief Communications Officer, announced 31st August and effective 1st September, reporting to chief executive Antonio Filosa. She joins from Johnson & Johnson, where she was global vice president of MedTech communications and public affairs.
• Talanx, the German insurance group, has moved group communications from Andreas Krosta, who departs on 31st August after 13 years, to Dr Martin Weldi, effective 1st August. Weldi has led group corporate development since 2018 and takes on communications in addition to that role; Bernd Sablowsky, head of investor relations and M&A, takes on external communications, reporting to Weldi.