Blog
Writing on writing (and AI)
Engineering, design, and product writing from the Inksong team.
What AI detection scores can and can't tell you
A follow-up to our heuristic-detector post. Three illustrative iteration examples showing what the score moves on — and what it doesn't.
Maintaining brand voice across a six-writer content team
Voice Profiles aren't only for individuals. Here's how a marketing team uses them to keep brand voice consistent across multiple writers.
How researchers humanize literature reviews without losing the citation graph
Researchers draft long lit-review prose with AI assistance — and need the citation structure to survive humanization intact. Here's the workflow.
The prompt strategy behind document-level humanization
Most paraphrasers work sentence by sentence. Coherent humanization needs cross-paragraph awareness. Here's how Inksong structures the Claude prompt.
NGN-first pricing in an English-language SaaS
We anchored pricing to Nigerian SaaS purchasing power, not USD conversion. Here's the math, the politics, and what we changed after launch.
Format preservation at scale: parsing DOCX without breaking the formatting
Humanized prose is only useful if it lands back in your workflow with the formatting intact. Here's the parser-rebuilder pipeline.
Building a content-aware diff for humanized text
A standard diff is line-oriented and useless for prose. Here's how we render before/after for documents users actually want to read.
Why we built a voice cloner instead of just a paraphraser
Paraphrasers homogenize prose. A good humanizer should sound like the writer who used it. Here's how Inksong's Voice Cloning works under the hood.
Honest about AI detectors: what our heuristic measures, and what it doesn't
Inksong shows a before-and-after AI-likelihood score. It's a heuristic — not a third-party guarantee. Here's exactly what's in it.