How to Remove AI Slop Without Erasing Your Voice
Remove generic AI-writing habits with a four-pass edit that protects facts and voice, names each problem, makes the smallest useful change, and ends with human review.
15 primary guides
Research, synthesis, drafting, editing, and source-verification workflows.
Main subject
These guides have Writing & Research as their primary topic.
Remove generic AI-writing habits with a four-pass edit that protects facts and voice, names each problem, makes the smallest useful change, and ends with human review.
Turn a messy read-later queue into a focused writing brief without letting AI flatten disagreement, detach highlights from context, or invent the missing evidence.
Start with an authorized recording, map the spoken and visual evidence, attach important claims to usable sources, and draft one focused article with its own point of view.
Audit old AI articles claim by claim, verify sources and dates, then decide what to patch, schedule, rewrite, merge, or retire without pretending automation can judge meaning.
Audit a thesis discussion with AI using source-linked claims, a worked example, design-specific checks, and prompts for careful revision.
Get comments tied to your rubric and draft, diagnose source and reasoning problems, and verify the revisions you write yourself.
Make page-cited flashcards from readable, scanned, visual, or long course PDFs, with worked examples, checks, and Anki or Quizlet import files.
Start with one checked passage, answer before seeing feedback, and use mistakes to choose changed questions and later review.
Evaluate document Q&A RAG with a runnable kit, a retained model trace, citation review, and questions the corpus cannot answer.
Compare an AI draft with the source passages you provide, catch unsupported numbers and overbroad wording, and leave each claim with a clear edit or verification step.
See when Search and follow-ups are enough, when a cited research report helps, and what one matched test can and cannot establish.
A practical guide to ChatGPT Deep Research: choose web, file, or app sources; write a useful prompt; edit the plan; monitor the run; verify citations; and export the report.
Follow a real ChatGPT Search answer through NASA source checks, then use the same claim-by-claim method for changing product facts and mixed claims.
Put two documents in one ChatGPT chat, request a source-backed comparison, verify important rows, and adapt the method for long, scanned, or consequential files.
Attach a PDF in ChatGPT, prompt for the summary you need, check page-level claims, and handle long, scanned, or visual documents.
Also relevant
These overlap with Writing & Research, but their main subject belongs elsewhere.
Turn a source-linked report into an accessible interactive web page without losing citations, denominators, limitations, or approval state. Follow a tested conversion with downloadable HTML and checks.
Why it fits here: The conversion protects citations and source-linked claims.
Turn an existing PowerPoint into a useful report without mistaking the slides for evidence. This workflow shows how to inspect the deck, reconnect original records, preserve uncertainty, and verify the finished document.
Why it fits here: The report traces claims in slides to the source material.
Write a performance self-assessment with ChatGPT using a complete work-log example, checked metrics, shared-credit guidance, and practical prompts.
Why it fits here: The self-assessment ties claims to entries in the work log.