(Source – Shutterstock)
Turnitin launches new AI writing detector for educators and universities
Turnitin, a leading provider of academic integrity solutions, has activated new features that detect the use of artificial intelligence (AI) writing tools.
According to the provider, these features can be found in Turnitin Feedback Studio (TFS), TFS with Originality, Turnitin Originality, Turnitin Similarity, Simcheck, Originality Check, and Originality Check+. The groundwork for developing these features started early — nearly two years before the release of ChatGPT.
“The technology behind ChatGPT, Google Bard, and other AI writing tools did not come as a surprise, given the dedicated research and development of the Turnitin AI team over the last few years,” said Chris Caren, CEO of Turnitin.
“However, we were all impressed with the progress and promise that GPT-3.5 showed over GPT-3.”
That progress even caught Elon Musk and other AI experts by surprise — so much so that they are calling for a six-month pause in developing systems more powerful than OpenAI’s newly launched GPT-4.
James Thorley, the Regional Vice President of Asia Pacific at Turnitin, explains that new features would give educators a powerful tool to maintain academic integrity across their student cohorts by providing powerful, actionable, and cross-institution data insights to enrich decision-making.
With this, over 10,700 institutions and more than 2.1 million educators can quickly evaluate a submission for the presence of AI-generated text and provide feedback to students in their current Turnitin workflows.
The work is not done
Generative AI tools, such as AI chatbots, are growing at a breakneck pace. Two months after OpenAI launched GPT-3.5 (more commonly known as ChatGPT), the chatbot had more than 30 million users and got roughly five million visits a day, two people with knowledge of these figures said.
Turnitin’s CEO knows this well and recognizes that new AI writing tools are coming out regularly with claims and aspirations of being undetectable.
Currently, the new features on this platform have reportedly been able to detect the presence of AI writing with 98% confidence and a less than one percent false-positive rate in Turnitin’s controlled lab environment.
Even then, it’s impossible to get a tool that can detect AI-generated text with 100% certainty. Muhammad Abdul-Mageed, a professor who oversees research in natural-language processing and machine learning at the University of British Columbia, explains why that’s the case.
“The whole point of AI language models is to generate fluent and human-seeming text, and the model is mimicking text created by humans,” Abdul-Mageed shares. “New AI language models are more powerful and better at generating even more fluent language, which quickly makes our existing detection tool kit outdated.”
Take GPT-4, for example. While it is less capable than humans in many real-world scenarios, the large multimodal model (accepting image and text inputs, giving text outputs) exhibits human-level performance on various professional and academic benchmarks.
GPT-4 passed a simulated bar exam with a score around the top 10% of test takers. In contrast, GPT-3.5’s score was around the bottom 10%.
Caren shares that Turnitin will continue to adapt and respond to the next iterations and innovations in AI writing — prioritizing the safety of students as well as the needs of educators and institutions.