You picked up three sections at $3,400 each in August, and two of them landed at 90 students because the department merged the under-enrolled ones. That’s 180 composition essays per assignment cycle, four cycles a semester, plus discussion boards nobody told you were graded. At fifteen minutes an essay you are working 180 uncompensated hours for a contract that pays like 45. Meanwhile your institution bought a Gradescope site license you were never trained on, or a Packback seat you were told to “just try,” and the 2026 pricing and feature changes at both vendors mean the advice you found in a Reddit thread from 2024 is now wrong in ways that will cost you a weekend.
This is written for working adjuncts and contingent faculty who already run a course in Canvas, Blackboard Ultra, or Moodle and can build a rubric without help. You should be comfortable clicking through an LMS settings panel and reading a pricing page critically. It is not a general “AI in education” overview, not a defense of or attack on automated assessment as a philosophy, and it does not cover K-12, proctoring tools, or plagiarism enforcement policy. If you want a tool decision you can defend to a department chair by Friday, you are the reader.
Be clear about what the machines are actually good at. AI grading is strong on structured, criterion-anchored work: code that either compiles and passes tests or doesn’t, short answers with defined correct forms, rubric-dimension consistency across a large pile, and first-pass feedback volume that no human can sustain at 90 students. It is weak — sometimes badly weak — on argument quality, voice, rhetorical risk-taking, and anything where a student is doing something interesting that the rubric didn’t anticipate. AI writing detection produces false positives on ESL and multilingual student prose at rates high enough that acting on a score alone is professionally reckless. Human review is non-negotiable on three things: any grade that affects a student’s standing, any flag that could become an academic integrity referral, and any borderline case near a letter-grade boundary. Both platforms in this guide will let you skip that review. Don’t.
What This Guide Covers
- A direct, evidence-based verdict on which platform fits which teaching load — with the cases where the answer is “neither”
- What actually changed at Gradescope under Turnitin ownership in 2026, and how it affects your grading workflow and your students’ data
- How rubric-based AI grading works under the hood, so you can predict where it will fail before it fails on your students
- A head-to-head scorecard across accuracy, setup time, faculty trust, accessibility compliance, and how hard it is to leave
- A realistic path to a working Gradescope rubric and code autograder inside a single 45-minute sitting
- How to stand up Packback for a large composition section without losing a week to configuration
- LMS integration through LTI 1.3 in Canvas, Blackboard Ultra, and Moodle when you have no admin rights and no IT ticket queue
- Per-course pricing math run against real adjunct section pay, including who pays when the department won’t
- Hours-saved benchmarks from two very different loads — a 90-student composition section and a 40-student intro CS course
- What FERPA actually requires of you, what these vendors do with student writing, and which contract terms to read before you upload anything
- How to handle AI-detection false positives on ESL writing without damaging a student or exposing yourself
- The failure modes that sink adoption: rubric drift across sections, appeals volume, regrade workflows, and the trust collapse that follows one bad automated grade
- Four case studies of adjuncts in four departments who reached four different conclusions — and the variables that drove each
- Fallback options worth knowing: SpeedGrader, CoGrader, EssayGrader, and FeedbackFruits, with the specific situation each one wins
Delivered as an instant digital download — you get full access immediately after checkout, readable on any device. One purchase, complete guide, no upsell, no subscription, no follow-on course.
Reviews
There are no reviews yet.