Nobody Reads Press Releases. AI Does.

AI has quietly become one of the most important audiences for press releases, and most comms teams don't know it yet. Data from 8 major LLMs shows that press releases rank second only to Wikipedia as a source AI draws from when defining brand narratives. That means errors in titles, roles, or company descriptions don't just disappear, they get embedded in AI systems and compound. The fix is straightforward: treat every press release as a structured input into AI, not just an announcement to journalists.

Here's something almost nobody in communications is talking about: press releases may be more important now than they've been in a decade. Not for the reasons they used to be. Not because journalists are reading them. Because AI is.

The Conventional Wisdom Is Wrong

The standard take in our industry right now goes something like this: press releases are corporate hygiene. They serve regulatory and investor relations functions. But for reputation building, they're a rounding error — wire distribution creates the appearance of coverage without the substance. Every marketing podcast has run some version of this argument.

We thought the same thing. Then we started tracking it.

What the Data Actually Shows

Through our AIQ platform, we track what 8 major LLMs — ChatGPT, Gemini, Perplexity, Claude, Copilot, Grok, AI Overview, and AI Mode — say about brands and their reputations. Press releases rank number two in terms of surprise factor, right behind Wikipedia, in how heavily AI models draw from them when constructing brand narratives.

People used to say 'who reads press releases?' AI does. Religiously.

ChatGPT in particular appears to favor PR Newswire content significantly more than other wire services. When a press release goes out on PR Newswire with accurate, well-structured information about a brand, executive appointments, product launches, market positioning — that content enters the information ecosystem that LLMs are continuously drawing from.

Why This Matters More Than You Think

If a press release gets your executive's title wrong, or forgets to mention their three most important prior roles, or describes the company in outdated terms — that error doesn't just live on an aggregator site nobody reads. It lives in AI. Potentially forever. Because press releases feed Wikipedia editors, who update articles, which feed LLM training data. The chain is real and it's fast.

The New Press Release Strategy

This changes how you should think about press release quality, structure, and content. A press release is no longer just an announcement to journalists. It's a structured data input into the systems that are defining your brand narrative for millions of users.

Key Principles for AI-Optimized Press Releases

Accuracy is non-negotiable — every title, every role, every company description needs to be current and correct. Entity clarity matters — use full names, official titles, and consistent company descriptions. Include the context AI needs — don't assume the reader knows your history. If your CEO joined from a Fortune 500 company after running a $2B division, that context should be in the release, because that's the context AI will use to describe them.

Structure the release for machine readability. Clear headers, declarative sentences, specific claims. The same characteristics that make a Wikipedia article effective make a press release effective for AI consumption.

What We Tell Our Clients

We now review press releases for AI readiness before they go out. It takes 15 minutes and it's become one of the highest-leverage activities we do. A single poorly structured press release can create narrative problems that take months to fix in AI systems. A well-structured one creates narrative assets that compound.

In a recent analysis we did for our client across their own corporate peers, the hierarchy of sources that AI trusts goes roughly like this: your own corporate site, Wikipedia, Earned media and press releases, LinkedIn, other bios, and surprisingly, other platforms like Medium and blogs. Understanding where press releases sit in that hierarchy — and treating them accordingly — is one of the simplest strategic shifts a communications team can make.

The death of the press release has been proclaimed annually for at least fifteen years. It keeps not dying. And now, quietly, it's become one of the most important inputs into the systems defining brand reputation for the AI era.

The audience changed. The medium didn't. Adjust accordingly.

AIQ tracks what 8 major LLMs say about your brand, including what sources they're drawing from. See a demo to find out where your press releases stand.