Slop*Detector

SUSPICIOUSLY VIBEY — slop score 73/100

The writing on this post carries several of the clearest LLM fingerprints we look for, and they show up in clusters rather than isolated instances. The em dash usage around the mason-jar and cervical mucus passages, paired with the rule-of-three list of 'rainwater, creek water, tap water,' reads like generated copy dressed up as personal essay. The repeated stacked-sentence drumbeat, lines like 'The claim? Period blood has healing properties. True.' and 'Science is not God. It is a tool. I love science. I am a doctor.,' plus the glued-on payoff sentence 'That last sentence is where wisdom lives,' push this well past coincidence. The page itself is just a standard Substack template with platform-default chrome, so the pill badges and glassmorphism findings carry little weight here and shouldn't be read as design evidence. Given how many severity-3 writing tells overlap in the same few paragraphs, a score of 73 feels right. This isn't a clean case, but it's also not a slam dunk for full AI generation, since long stretches of the piece still carry a distinct, uncontracted personal voice that a model would struggle to sustain this long.

Findings

• Pill tags — Little fully-rounded pill badges sprinkled across the page ("New", "Beta", category chips) are a default garnish of vibe-coded UI.
• Untouched shadcn/Tailwind defaults — shadcn components, lucide icons, and Inter/Geist/Space-Grotesk font stacks left at their defaults mean the model picked the design, not a person. Instrument Serif as an accent font is a separate, independently corroborated tell.
• Glassmorphism — Frosted-glass backdrop-blur panels are a recurring generated-UI decoration.
• No notable layout/styling tells present — This is a standard Substack/newsletter template (masthead, subscribe button, author byline, like/comment/restack icons, inline email subscribe box) — this is platform-default chrome, not an AI-generated design decision, and the actual content area is plain long-form prose with embedded images and a pull-quote style bold text, which is consistent with human blogging on a hosted platform rather than a custom-built vibe-coded site.
• Embedded social media screenshot as content image — The TikTok/Reels screenshot embedded mid-article (jar with plant, heart/play icons, 60.1K likes) is a legitimate content image referencing real social media, not an AI-generated illustration — included here only to note it is not a vibe-coding tell despite being an unusual image.
• Em dashes — Em dashes are the single most reliable punctuation tell of LLM copy. Human marketing copy almost never uses them; models reach for them constantly.
• One/two-word punch sentences — Single dramatic words dropped as their own sentence ("Done." "Gone." "Period.") are an LLM emphasis habit, not a human one.
• Stacked short-sentence drumbeat — Three or more clipped sentences in a row ("Important emails buried. Drafts that sound wrong. Labels everywhere.") is the manufactured-rhythm tell.
• "That’s where/why/the…" payoffs — Short demonstrative payoff sentences glued onto paragraph ends ("That’s the flywheel.") are choppy LLM theater.
• AI vocabulary — Words statistically overrepresented in generated text: "seamless", "leverage", "unlock", "empower", "elevate", "robust" and friends.
• Compulsive rule-of-three — "Fast, reliable, and secure." Triads everywhere read as machine rhythm — humans vary list length; models default to three.
• Stiff, uncontracted prose — Expanded forms ("it is", "you will", "do not") dominating over contractions reads as robotic formality — either legal text or a language model.
• Em dash usage — Several em dashes appear mid-sentence for dramatic pauses or reframes, a pattern the guide flags as a classic LLM punctuation tell, though here it reads as part of an established personal voice rather than pervasive generation.
• 'It's not X, it's Y' reframe — The 'Science is not God. It is a tool.' construction is the textbook LLM reframe-for-emphasis move, even though it appears only once rather than as a running tic.
• Stacked parallel short declaratives — Clusters of short parallel sentences used for rhythmic punch resemble the drumbeat pattern the guide calls out, though they're not pervasive across the whole piece.
• 'That's where...' tacked-on payoff — A short demonstrative sentence glued onto a paragraph to deliver the 'point' is the exact structure the guide identifies as an AI tell, used once near the grimoire section.
• 'Genuinely' / 'real' as intensifiers — The guide flags 'genuinely' and 'real' used to manufacture sincerity rather than demonstrate it; both appear a handful of times but are not overused.