Write it again, better.
Folio gives English learners writing feedback at their CEFR level. AI drafts the feedback, your teachers own it, and every learner gets a real revision loop.
AI-drafted feedback · reviewed by the teacher
B1“In my country is very common to eat dinner late, even the children.”
Good detail here, and your time phrases are working well. One pattern to work on: English needs a subject in this position. Try “In my country, itis very common”. Find 1 more sentence like this in your draft, then revise.
Generic AI feedback became free. The right feedback didn’t.
Google Classroom now drafts essay comments at no cost. But a tool that can’t tell a B1 writer from a C1 writer gives both the same feedback, pitched at neither. Meanwhile, the AI detectors that promised integrity flagged 61% of essays by non-native English writers as AI-generated.
Folio is built for the part that never got commoditized: feedback that’s right for a particular learner, with a teacher behind every word. Our research review explains why.
One loop: draft, feedback, revise
Write, get feedback, revise, resubmit. It’s the pedagogy every writing teacher believes in, and few have time to run for 120 students. Folio makes the revision loop the default assignment structure: learners draft, get feedback at their level, and draft again. An AI coach scaffolds their thinking along the way and never writes the essay for them.
Your teachers stay in charge
AI drafts feedback and suggests rubric scores across 11 CEFR criteria, from task achievement to mechanics, each with its own confidence. The teacher reviews, overrides, and decides what learners see. The teacher’s name is on the feedback because the teacher owns it.
Feedback has a level
A B1 learner and a C1 learner should not get the same comment. Folio profiles each draft for the vocabulary and grammar structures the learner attempted, then pitches feedback at their CEFR level: what it targets, and how it’s worded. Whether the writing achieved the task counts too, not just the grammar.
We show the work, not a verdict
Folio keeps draft and revision history that a learner and teacher can look at together. There is no AI-probability score, and there never will be: detectors misread exactly the writers we serve. A draft history is evidence. A percentage is an accusation.
Watch mastery grow
Learners see the structures and vocabulary they’ve mastered grow over time, from attempted to accurate. Badges mark real milestones, the portfolio keeps every piece, and error patterns are something to work on, not something to hide. Progress is visible, so the next draft has a reason.
What we hold ourselves to
These commitments come from our public research, and we expect to be held to them.
- A teacher reviews AI feedback before students see it.
- Feedback quotes the student's actual writing.
- Feedback is pitched at the learner's level and covers the task, not just the grammar.
- Provenance, never verdicts. No AI-probability scores, no keystroke surveillance.
- Growth is the point. Revision is the default assignment structure.
- We'll show our work: our scoring evaluation gets published, agreement rates included.
- Student writing is not training data.
From the lab
All researchBring Folio to your program
We’re piloting with language programs now. Tell us about your learners and we’ll set up a walkthrough with your teachers.