Press Releases in the Age of AI: Writing B2B Tech News for Journalists and Algorithms

"Are press releases dead?" is a question that gets asked every year, and the answer is always the same: no—but the bad ones should be. In 2026, press releases serve a fundamentally different purpose than they did even five years ago. They're not just for journalists anymore—they're for AI systems that index, interpret, and cite them.

The Dual Audience Problem

Every B2B tech press release now has two audiences: the journalists you're trying to inform (and ideally inspire to write a story) and the AI systems that will process, index, and potentially cite the release in generated answers.

These audiences have different needs. Journalists want compelling angles, quotable executives, and exclusive access. AI systems want structured data, clear entity associations, and factual claims supported by evidence.

Writing for Journalists in 2026

• Lead with the "so what" — Don't bury the news. The first sentence should tell a journalist why their readers care.
• Include real data — Revenue growth, customer counts, performance benchmarks. Journalists need proof points that make the story defensible.
• Provide quotable executives — Generic quotes like "We're thrilled to announce..." waste everyone's time. Give journalists quotes they can actually use.
• Offer exclusive angles — Signal that you're open to exclusives, interviews, and deeper access. The press release is the starting point, not the destination.

Writing for AI Systems

Entity Clarity

LLMs need to understand who you are, what you do, and how you relate to other entities (competitors, partners, categories). Be explicit. Don't assume the AI knows your company—define it clearly in every release.

Structured Claims

AI systems extract factual claims. Make your key claims explicit and evidence-backed: "Company X, a leader in [category], today announced [specific news] that enables [specific outcome]." This structure helps LLMs accurately represent your announcements.

Category Association

Repeatedly associate your company with your target category. If you want LLMs to surface you when asked about "best cybersecurity platforms for healthcare," your press releases need to consistently connect your brand to those terms.

Boilerplate Optimization

Your company boilerplate appears on every press release. Optimize it for AI discoverability: include your category, key differentiators, notable customers, and the specific problems you solve. This boilerplate is training data for LLMs.

Distribution Strategy

Wire distribution (PR Newswire, Business Wire) still matters for reach and indexing, but it's not enough. Complement wire distribution with:

• Targeted, personalized pitches to key journalists
• Social media amplification with executive commentary
• Posting to your website newsroom with proper schema markup
• Sharing with analysts who track your category

At Crackle PR, every press release is engineered for dual impact—crafted to compel journalists and structured for AI systems using our AEO News Release framework. Combined with strategic media relations, this approach ensures your announcements work for both audiences.

Press releases aren't dead—but the ones that ignore AI are writing for half their audience. In 2026, every announcement should be built for both journalists and algorithms.