The rise of AI media training

Andrew Petro is account director at Matter Communications. 

Journalists are receiving more pitches, email interviews, and contributed content than ever before, and much of it sounds the same. The language is polished, cautious, and interchangeable. That sameness has become a red flag. Many publications now discourage or outright prohibit AI-generated submissions, and some reporters, including tech journalist Maria Korolov, have said they prefer live interviews because written responses increasingly read like they were produced by a machine.

In a media environment where reporters are actively looking for real expertise, that loss of distinctiveness can quickly cost you credibility. That is why AI training is emerging as a practical supplement to traditional media training. Spokespeople now need to understand how AI can subtly shape their voice and dilute their perspective. The risk is not using AI. It is relying on it so heavily that answers start to sound generic.

How to train spokespeople for an AI-saturated media environment

AI training works best when it feels practical, not philosophical. Executives do not need another lecture about responsible AI use. They need to see, in real time, how easily their expertise can become generic and how to prevent that from happening. Below are four practical ways to weave this concept into the standard media training playbook:

• Run an AI comparison drill

One of the most effective exercises is the “AI versus you” drill. Start with a real reporter question relevant to the executive’s role. For example, imagine the CEO of a cybersecurity company is asked, “How are your enterprise customers adjusting their security investments in this market?” Before the session, generate a standard AI-style answer to that question. It will likely mention macroeconomic uncertainty, digital transformation, evolving threat landscapes  and the need for resilience. It will be technically correct and completely interchangeable.

Then ask the CEO to answer the same question without notes. Push them to reference an actual customer conversation from the past quarter, a deal that almost stalled, or a board discussion about shifting budget priorities. When you compare the two answers side by side, the difference is obvious. The AI version describes a generic trend. The executive version explains what actually happened and why it mattered. The contrast is immediate and often eye-opening.

• Practice specificity

AI defaults to abstraction, but strong interviews depend on detail. In mock interview scenarios, I will often stop an answer midway and ask follow-up questions that a reporter would ask. If a chief revenue officer at reverse logistics company says, “We’re seeing increased demand from enterprises,” the immediate response in training could be, “Increased compared to when? From which industries? What’s driving demand and why?”

The goal is to train leaders to anticipate those questions and tighten their answers before a reporter has to pull the specifics out of them. Over time, they begin to replace general statements with sharper ones on their own.

• Build opinion discipline

Another critical component is opinion discipline. AI-generated content often avoids taking a stance. Media training should encourage spokespeople to articulate what they believe and why, even when the answer is nuanced. Clear thinking is more valuable than safe language.

In a session with the head of product at an AI company, for example, we might practice answering questions like, “Will AI replace jobs?” The first instinct is often to hedge. Instead, we work on articulating a clear position, even if it is nuanced.

Perhaps the answer is acknowledging that yes, AI will replace some jobs, but it also has the potential to create a wide array of new high-quality job opportunities. Reporters are far more likely to use a quote that takes a thoughtful position than one that sounds like it came from an approved messaging doc.

• Normalize natural language

Finally, AI training should normalize imperfection. Slightly informal phrasing, pauses, or course corrections signal authenticity. Polished but generic language does not. Helping leaders get comfortable with sounding human is one of the most valuable outcomes of this training and can go hand-in-hand with your traditional playbook.

During mock interviews, we record short segments and play them back. More often than not, the answer that felt less rehearsed is the one that sounds more credible. In a media environment where reporters are actively screening for generic responses, that authenticity becomes a competitive advantage.

The future of media training is human-first

The goal of AI training is not to make spokespeople sound smarter or faster. It is to help them preserve what makes them valuable in the first place.

Traditional media training still matters, but it is no longer sufficient on its own. In a media environment flooded with synthetic content, credibility comes from specificity, experience and opinion. AI training is the layer that helps spokespeople protect those qualities and stand out for the right reasons.