Top perspectives

  • Rico BolosScaled a creator program to 450+ at a stealth AI startup
    Product Generalist at heyclicky

    I got a PM take-home once that banned AI on my end, then the feedback came back reading like ChatGPT wrote it. Good style still has a point of view; what’s happening now is mostly people reaching for the same safe shape because they think that’s what passes.

  • Robyn AndersonNational AI speaker for SCORE
    Founder of Jam AITrained 5,000+ people in AISold a $575,000 website

    Most take-home work now reads like it came from the same intern because people prompt for professional and accept the first draft. I make my people paste real emails and old captions first, then delete the AI giveaways like in today's fast-paced world before anything goes out.

  • Mandy McEwenBlames lazy prompts for generic AI content
    LinkedIn, Personal Branding, and Sales Navigator Instructor at PavilionLinkedIn Top 24 B2B MarketerSocial Media Marketing World 2022 speaker

    Most take-homes look the same because people are prompting AI with zero context and calling it done. I’ve been writing short, spaced-out posts for ten years, and now that same formatting gets flagged as AI, so the real signal is whether there is a lived opinion underneath the style.

  • Eric AndalWarns AI‑generated posts erase authentic voice
    I ghostwrite LinkedIn posts for AI Executives.Published EEC Starter Kit

    Most take-homes look the same because people are handing AI a blank prompt and it reaches for the median, the safest word in every slot. Good style has fingerprints; AI slop has none.

  • Julia FedorinCo‑built Echo AI video editor
    Video Producer + Creative Lead at Composio

    Mostly AI, and the giveaway is the polish. I can spot it in scripts now: sentences that are too clean, titles with every word capitalized, and conclusions that wrap up too neatly. After three months of writing with Claude and GPT, I started doing the same thing, and it made my skin crawl.

  • Kamil BancFractional Chief AI Officer at Kwik Brain
    Pre-sold $1M in fragrancePanel at Jim Kwik's Limitless Live

    It's both, and that's the trap. I rewrote job postings with a prompt and watched the same strong leadership, cross-functional collaboration shape appear everywhere, because AI collapses toward the high-probability answer, and the high-probability answer now looks like good style. The real signal is whether the submission shows a process rebuilt around AI, not whether it reads clean.

  • Saeid AmanzadehSenior Software Developer at Pearson
    Software Developer at SefrYek

    Most of the sameness I see in submissions now is the same polished structure everyone learned from the same AI tools: clean intro, bullet points, neat conclusion. Good style has always been consistent, but it used to carry a person's actual decisions and trade-offs, not just the shape of them.

  • Gillian GontardAI Evaluator at Handshake
    BFA Printmaking, Corcoran at GWUAI Evaluator at DataAnnotation

    I spent years grading student work and now evaluate AI outputs for a living, and the sameness you are seeing is mostly the same thing by a different route. Good style has always converged on what gets rewarded, but a model can now produce that convergence on demand, so the line between imitation and competence has gotten very thin.

  • Darren Petronella, MS, CPRWSenior Adjunct Professor at Adelphi University
    Professor/Career & Transfer Counselor at Nassau Community CollegeOwner/Career Counselor/Coach of Career Pivot 123 LLC

    Most of the sameness I see is AI flattening the voice out. A strong take-home still has the candidate's fingerprints on it, and that is what I am looking for.