‘Anti-Grammarly’ AI tool adds typos on purpose—what it means for business email
anti-Grammarly AI – An AI “anti-Grammarly” extension, Sinceerly, deliberately inserts typos to make emails feel human—raising questions about trust, engagement, and what executives actually want.
AI writing has made speed and polish the default. Now one new browser extension is doing the opposite: it deliberately nudges emails toward “imperfection” so they sound less like a machine.
The tool. called Sinceerly. has drawn attention for its ability to “un-AI” a message—complete with misspellings and informal phrasing—built by venture investor Ben Horwitz.. The focus keyphrase here is “anti-Grammarly AI. ” and its premise lands with a direct business question: if recipients can sense AI. can controlled imperfections become a competitive advantage?
Sinceerly is positioned as a satirical “anti-Grammarly” extension, mirroring Grammarly’s general interface but changing the outcome.. Users can choose among three editing levels—subtle. human. and “CEO.” The “subtle” setting aims to shorten and simplify AI-heavy paragraphs.. The “human” mode leans further into casual language, including slang and abbreviations.. “CEO” goes the furthest. transforming messages into brief. rough drafts that resemble the kind of note busy executives might send after skimming what they received.
That design choice matters because it targets something businesses worry about right now: the growing sameness of AI-generated communication.. Many inboxes are crowded with emails that read like they were produced by the same template—confident tone. clean structure. and just enough generic phrasing to avoid risk.. In that environment, even small deviations can act like a signal.. Sinceerly tries to manufacture that signal intentionally, turning what usually counts as “mistakes” into a recognizable style.
Horwitz has described the project as a “hold up a mirror” to our complicated relationship with technology—especially the tension between using AI to write and wanting to appear authentically human.. While the product is framed as humor. the underlying mechanics are straightforward: the extension rewrites AI-like text and compresses it. then increases roughness as the user selects higher intensity levels.
The commercial curiosity is obvious, too.. Sinceerly has been tested by sending emails to major corporate leaders—reportedly with responses from several of them.. Horwitz also shared the tool’s approach publicly. and the trial results quickly became part of the broader conversation: not just whether typos get attention. but whether recipients interpret them as evidence of a real person behind the message.
From a business perspective, that interpretation is where the real value could sit—or where the risk begins.. A typo might increase perceived humanity for some readers. but it can also reduce trust. especially in industries where accuracy is treated as a baseline competence.. There’s also a brand-safety question: “human-like” language can be misread as careless. and in regulated or high-stakes contexts. the cost of seeming sloppy can outweigh any engagement gains.
At the same time, marketing experiments already suggest that perception often beats polish.. The key idea isn’t that errors are inherently good—it’s that audiences react to signals of authorship.. If a message appears human. recipients may be more willing to open. respond. or at least give it the benefit of the doubt.. Sinceerly effectively tries to exploit that psychological shortcut by converting the visible artifacts of imperfection into a consistent pattern.
That’s also why the tool’s most interesting contradiction may be strategic rather than comedic.. Horwitz used AI systems to build an extension meant to make writing look less like AI.. For businesses. the takeaway is not “stop using AI.” It’s a more operational lesson: teams are under pressure to move fast. and AI is currently one of the quickest ways to move from idea to output.. The more difficult work is translating that output into something that fits a specific relationship—one that feels credible in context. not merely fluent on the page.
Even the naming and rollout reflect a practical reality: attention is currently the scarcest resource.. Sinceerly’s viral pull suggests people are tired of perfectly optimized messaging and want contact that feels more lived-in.. Yet for any company considering tools like this. the smartest approach may be selective use—treating imperfection as a stylistic dial. not a blanket policy.
Misryoum view: the rise of anti-polish tools signals a shift in communication strategy.. As AI writing becomes routine, differentiation may move from grammar quality to human signaling—tone, brevity, and authorial fingerprints.. The “anti-Grammarly AI” trend could become less of a novelty if it helps businesses get replies without sacrificing credibility. but only if teams define boundaries for when imperfection helps and when it damages trust.