What AI Actually Changes in Communications: Operations, Not Creativity

The communications industry's relationship with AI has been defined by anxiety about the wrong things. The dominant fear — that AI will replace the creative and relational work that practitioners do — has produced a defensive conversation that misses where the actual disruption is happening.

AI is not, in the near term, replacing the judgment that makes a great communications professional. It is not replacing the relationships, the strategic thinking, the craft of the pitch, the ability to read a room and understand what a journalist needs. Those remain human.

What AI is doing — and doing right now, in ways that are already reshaping workflows — is attacking the operational layer that sits beneath the craft. The research, the monitoring, the drafting, the synthesis, the pattern recognition across large amounts of information. This is where the real change is happening. And it is significant.

What generative AI actually did

When large language models became capable enough to produce coherent text at scale, the communications industry experienced an initial wave of enthusiasm followed by a more complex reality. The enthusiasm was about content creation: if AI can write, maybe it can write press releases, pitch emails, blog posts, social content — all the production work that consumes significant practitioner time.

91% of communications professionals report using generative AI tools in their work. But 50% also cite AI as a top challenge — a sign that the adoption is happening faster than the workflows to support it.

The reality is more nuanced. AI-generated content, without significant human editing and strategic direction, tends to be generic. It can produce a competent press release on a brief, but competent press releases were never the bottleneck. The bottleneck was always the judgment that determines what angle to take, what to emphasize, what the journalist actually cares about, how to position the client authentically in a competitive landscape. AI does not yet provide that.

But AI does something else that is less celebrated and more valuable: it dramatically accelerates the work that precedes the creative work. The research. The synthesis. The pattern recognition. The monitoring and analysis. These are the tasks where AI is genuinely transformative — not because it eliminates the need for human judgment, but because it compresses the time and effort required to build the foundation that human judgment operates on.

The messy reality of AI adoption in PR

The current state of AI adoption in communications is characterized by high enthusiasm and low integration. Most practitioners have experimented with AI tools — primarily ChatGPT and similar LLMs — for specific tasks. They use them to draft, to brainstorm, to summarize, to generate options. The tools are helpful for these tasks, used episodically.

What has not happened yet, for most teams, is the integration of AI into the operational workflows that determine how work gets done day-to-day. The research process is still largely manual. The monitoring is still largely manual. The synthesis of what's happening in a client's media landscape — the regular process of reading coverage, identifying patterns, assessing what's changing — is still done by humans, slowly, incompletely.

The gap is not between teams that use AI and teams that don't. It is between teams that use AI as a collection of tools and teams that use AI as infrastructure — something that runs continuously, processes information at scale, and delivers organized intelligence rather than raw output.

That gap is widening. And teams that are still treating AI as a collection of discrete tools — useful for specific tasks, but not integrated into how work flows — are going to find themselves at an increasing disadvantage relative to teams that have figured out how to build AI into the operational layer.

Tools versus infrastructure

The distinction between AI as tool and AI as infrastructure is not semantic. It describes a fundamentally different relationship with the technology.

AI as tool: a practitioner has a task, they use an AI tool to help with that task, they review the output and use what's valuable. The tool is passive — it responds to requests. The workflow is still organized around human initiation and review.

AI as infrastructure: the system runs continuously, processes incoming information — new coverage, journalist activity, competitive moves, trend signals — and surfaces what matters. The practitioner doesn't have to initiate the research; the research is already done. The intelligence is available when they need it. The workflow is organized around human judgment operating on pre-processed, organized information.

The difference in practical output is substantial. A team working with AI infrastructure has a continuously updated model of their clients' media landscape, their category narrative, the relevant journalist interests, and the emerging opportunities. A team working with AI tools has faster first drafts.

Redrawing the boundary

What AI is doing, at its most significant, is redrawing the boundary between what requires human judgment and what doesn't. Research that previously required human hours — reading articles, tracking patterns, building journalist profiles, monitoring competitive coverage — can now be done by systems operating at machine scale and speed.

This does not mean that human judgment matters less. It means that human judgment can operate on better, more complete, more current information. The strategic question — what angle to take, which journalists to target, how to position the client in a specific moment — still requires human experience and craft. But the foundation that question is answered on can be dramatically richer.

The practitioners who understand this — who are building their workflows around AI infrastructure rather than AI tools — are going to produce better work, faster, with less operational friction. The ones who are waiting for AI to replace the judgment work, or who are using AI only for first drafts, are missing where the leverage actually is.

The transition is not finished. The tools are still developing. The workflows are still being invented. But the direction is clear: AI in communications is an operational transformation, not a creative one. The creative and relational work remains human. The infrastructure that supports it is changing rapidly.