A single sentence construction has become so common in AI-generated writing that it has effectively become a fingerprint. The pattern "It's not just this, it's that"has more than quadrupled in corporate communications since 2023, according to a new analysis, raising uncomfortable questions about how much of what major companies publish is now being written by machines.
The Data Is Striking
A Barron's report used market intelligence firm AlphaSense's database to scan corporate news releases, earnings reports, and government filings for the sentence pattern. The results were dramatic: usage jumped from roughly 50 mentions in 2023 to over 200 in 2025 a more than fourfold increase that tracks almost perfectly with the rise of generative AI adoption in corporate communications.
The pattern works by presenting an idea and then immediately escalating it. "It's not just a tool it's a collaborator." "It's not just on the horizon it's already unfolding." The construction sounds authoritative and confident, which is exactly what corporate communications teams want. It is also exactly what language models tend to produce when asked to write in a professional, persuasive tone.
Big Tech Is Doing It Too
The trend is not limited to obscure press releases. Some of the world's largest technology companies have published communications saturated with the pattern. Cisco, Accenture, McKinsey, and Workday have all used it in recent blog posts and trend reports.
Microsoft's communications have been particularly notable. In a single blog post by CEO Satya Nadella, the pattern appeared multiple times describing the company's founding vision, its approach to AI tools, and its ambitions for making AI accessible to everyone.
Whether these specific communications were AI-assisted is impossible to confirm. But the sheer concentration of the pattern in corporate output that did not contain it just two years ago suggests that generative AI is playing a significant role in how major companies now draft their public-facing content.
Why AI Writes This Way
The pattern is not random. AI models are trained on vast amounts of human writing, and the "not just this it's that" construction is a rhetorical device that humans have used for decades in persuasive writing, speeches, and advertising. What makes it distinctive in the AI era is its frequency and ubiquity.
When a language model generates text, it selects the most statistically probable next word based on the patterns in its training data. Persuasive corporate language heavily features this escalation pattern, so AI models reproduce it at a much higher rate than human writers naturally would especially when prompted to write in a polished, professional style.
The pattern has become so associated with AI-generated text that it now functions as a detection signal. Alongside other tells like excessive use of em-dashes, the word "delve," and overly balanced paragraph structures the "not just this, it's that" construction has become one of the most reliable indicators that text was produced or heavily assisted by AI.
The Bigger Problem
The trend reveals something deeper than a stylistic quirk. It suggests that a growing percentage of corporate communication the press releases, earnings narratives, blog posts, and reports that investors, journalists, and the public rely on to understand what companies are doing is being generated or substantially drafted by AI tools.
For investors, this raises questions about authenticity. If an earnings report's narrative framing was written by an AI, does it reflect genuine executive thinking or just the most statistically probable language? For journalists, it complicates source evaluation. And for the general public, it adds another layer of uncertainty to an information environment already struggling with trust issues.
A Self-Fulfilling Loop
There is also an ironic feedback loop at work. AI models were trained on human writing that occasionally used this construction. The models amplified it by using it far more frequently than humans ever did. Now, as AI-generated corporate communications flood the internet, they become training data for future AI models which will use the pattern even more. The construction becomes more common not because humans chose it, but because machines did.
The pattern may eventually fade as AI writing tools become more sophisticated and as companies learn to edit AI output more carefully. But for now, it stands as one of the clearest visible markers of how deeply AI tools have embedded themselves into the corporate communications pipeline whether companies want to admit it or not.
