From the Turnitin Product Team:
Today, Turnitin is expanding our AI writing detection capabilities to include AI bypasser detection. This enhancement will help educators identify likely AI-generated text even if it's been modified by an AI bypasser/humaniser, enabling them to uphold academic integrity in the face of this emerging challenge. These latest updates further ensure that Turnitin keeps pace with the evolving LLM landscape. 

This feature will automatically be available to all customers who currently license Turnitin Originality and have the AI writing detection capabilities enabled on their account.

It will also be available to iThenticate 2.0 customers who currently license AI writing detection and have it enabled on their account. This feature will only be available for English submissions.

Why this matters:

• Deter potential AI misconduct:  Our enhanced AI writing detection can now identify text that was altered by an AI bypasser to make it sound more human and evade detection. This capability helps deter students from using AI writing tools and potential misconduct.

• Draw actionable insights: Use the data and insights provided by our overall AI writing detection capabilities to start a formative discussion with students. The AI writing report highlights the text segments that our model predicts were likely written by an AI tool, including content that may have been further modified to avoid detection. Color-coded insights enable users to interpret results easily. 

• Support your existing workflow: AI bypasser detection is part of our AI writing detection capabilities, which are built into our existing Similarity Report workflow. This means it’s already integrated with the learning management system you’re using. You can check student papers for text that is likely AI-written, including AI-bypassed content, as well as AI paraphrased content, without changing the way you work.

With this update, we can also detect content generated by GPT-5. While our AI writing detection capabilities are not specifically trained on GPT-5, our early internal tests on 100% AI-generated content from GPT-5 indicate that our AI writing detection capabilities are able to identify content likely generated by this LLM.

Learn more in our product update and check out today’s blog post from our Chief Product Officer, Annie Chechitelli: Navigating AI misuse: Tackling the impact of AI bypassers on academic integrity
 

What do you think about this update? How might it change your workflow or classroom discussions? Comment below!