Slop*Detector

SUSPICIOUSLY VIBEY — slop score 52/100

This one lands as suspiciously vibey mostly because the underlying site is real and the tells sit on top of it rather than running through it. Deloitte.com's structure, mega-nav, color-coded section banners, and legal footer all check out as the work of an actual enterprise design team, not a template. What drags the score up is the Privacy Policy copy, which reads like it went straight from a draft to production with em dashes and phrases like 'protect what matters, your data, your customers, your future' never touched by an editor, plus a hero that leans on stacked gradient backgrounds and a 'from the mountains of Milano Cortina to the beaches of Los Angeles' sweep. Scattered AI vocabulary ('utilize,' 'streamline') and a tacked-on payoff sentence add minor noise but wouldn't move the needle alone. The verdict here is less 'this site was vibe coded' and more 'a legitimate site let an unreviewed AI draft slip into a legal page,' which is a real finding but a different kind of problem than a from-scratch slop site.'

Findings

• Feature-tile grid — Equal-width card grids of 3–4 identical tiles (icon, subheader, blurb) are the signature layout block of generated landing pages.
• Pill tags — Little fully-rounded pill badges sprinkled across the page ("New", "Beta", category chips) are a default garnish of vibe-coded UI.
• Gradient backgrounds / gradient text — Large gradient washes and gradient-clipped headline text (usually blue→purple) are the default decoration of generated UI.
• Real corporate site, not template scaffold — This is the actual Deloitte.com homepage: consistent brand system (Deloitte green/blue/teal color-coded section banners), custom photography, and editorial content structure typical of a large enterprise site built by a professional design/dev team over years.
• Card grid under 'The latest from Deloitte' — A 4-card row does technically match the feature-tile shape (image, headline, blurb, byline), but each card contains genuine distinct editorial content rather than generic templated feature copy, so this reads as a real content module, not a stock AI feature grid.
• Cookie consent modal (standard enterprise compliance, not a tell) — The cookie banner is lengthy and formal but this is standard legal/compliance boilerplate for a Fortune 500 company, not AI-generated filler; it includes real granular consent options and a link to an actual Cookie Notice.
• Legitimate legal footer — The footer contains fully fleshed-out legal language (DTTL structure disclaimer, Terms of Use, Privacy, Data Privacy Framework, Cookie Notice, Labor Condition Applications) with no placeholder text, consistent with a real corporate legal team.
• "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.
• From X to Y sweep — The hero copy uses the classic sweeping-range construction the guide flags as a tell, pairing two disparate locations to sound expansive rather than specific.
• Stacked short declarative sentences — Three short parallel sentences in a row build a rhythmic drumbeat effect at the open of the hero, though it's a single instance rather than a pervasive pattern.
• AI vocabulary word ('utilize') — 'Utilize' is on the banned AI-vocabulary list; plain 'use' would read more human, though this is a single occurrence in otherwise dense, specific business copy.
• Rule-of-three triad — A triad adjective stack appears in the USGA blurb, a minor instance of the compulsive rule-of-three pattern.
• Privacy Policy reads as generated — The Privacy Policy carries multiple AI-writing tells (em dashes, ai vocabulary). Generated legal pages are a strong marker of an unreviewed, vibe-coded ship.