Second Rater notes #1: an AI scorer can agree with examiners on average and still be wrong in one predictable place.
In this illustrative Spot-Check (fictional app, synthetic data), the scorer matched a blind second rater on 11 of 20 workplace emails. Seven of the nine...
An AI English tutor can sound convincing. But how do you check whether its feedback is consistent and appropriate for the learner?
I’ve launched Abdelaty Tutor Assurance Studio (ATAS), a practical toolkit for people building, selecting, or evaluating AI English speaking tutors.
It...
After a speaking mock, “be more fluent” is difficult to practise.
A useful next step is specific: identify the moment, explain what happened, and rehearse a better response.
My OET Medicine Live Mock + Personal Feedback Report includes a 45-minute session with two timed...
A polished explanation can still misdiagnose a learner’s writing.
Illustrative QA case — synthetic content, not a real learner or vendor result.
The learner writes: “I look forward to hearing from you.”
An AI reviewer calls “hearing” a grammar error and changes it to “hear”. The...
contra.com
AI Writing Assessment and Feedback Quality Audit by Suliman Abdelaty
Read more about AI Writing Assessment and Feedback Quality Audit by Suliman Abdelaty on Contra.
“AI allowed” is a starting point. The project still needs an agreed approach.
Three questions make the brief more useful:
Is generative AI excluded from deliverables?
Is assistance allowed while the work remains human-authored?
Are generated assets allowed, with each asset...
