Because of the high demand for software that detects AI-generated content, plagiarism software is increasingly outfitted with AI detection tools. However, platforms that detect plagiarism are often distinct from those that flag AI-generated content.
Plagiarism software has been around for decades, with Turnitin among the most well-known plagiarism-detecting products. AI detection tools are far younger, dating back to the early 2020s, when tools like ChatGPT became as common on students’ laptops as Spotify and iMessage.
AI detection tools and plagiarism checkers are increasingly being rolled into a single software package (or at least packaged together). However, these tools have some key differences—including the propensity for AI detectors to make inaccurate judgments and raise false flags.
These tools are alike in one key way, though. Their findings often serve as the impetus for misconduct allegations against college students. If you or your student is facing possible discipline for alleged academic misconduct, call the LLF National Law Firm Student Defense Team today at 888-535-3686 or contact us online as soon as possible.
What Is Plagiarism Software?
Plagiarism software is a class of programs designed to detect plagiarism within a submitted work. Platforms like Turnitin have served as a filter for professors for more than a quarter-century, identifying:
-
Entire essays copied and pasted verbatim from the internet
-
Patchwork plagiarism, in which the person has copied and pasted significant portions of their work from third-party sources
-
Uncredited quotations and ideas
-
Works that have been previously submitted for credit (identified by scouring an extensive database of previously submitted work)
One definition of plagiarism is “Presenting work or ideas from another source as your own, with or without consent of the original author, by incorporating it into your work without full acknowledgement.”
Before the widespread adoption of AI, this definition typically referred to other humans’ works being co-opted without adequate crediting of the source. Traditional plagiarism software was designed to catch such instances of traditional academic misconduct—one human copying another’s work and passing it off as their own.
What Is an AI Detection Tool?
AI detection tools (or AI detectors) are primarily designed to determine if a person (in this case, a student) has used generative AI to create a work. You could classify the use of AI to generate academic work as a form of plagiarism, at least based on the definition we previously cited—“Presenting work or ideas from another source as your own…”
Rather than “another source” being another human’s words or ideas, in this case, the other “source” is generative artificial intelligence.
AI detectors are meant to identify work generated by artificial intelligence, rather than by a living, breathing person. These detectors, which are known by names like GPTZero and Originality.ai, uncover AI-generated content by analyzing:
-
The predictability of the writing, as common patterns can be an indication of AI-generated work
-
The tone of the writing, as AI tends to write in certain tones that an AI detector picks up on
-
A repository of work created by AI, as AI-powered platforms can spit out words and phrases that are identical to its previous creations
-
The variation of sentence structure and length
-
Other features that might indicate a work is AI-generated
Like a poker player, AI has tells when it is creating written content. AI detectors are trained to read those tells and raise an alert when they determine some or all of a given work has been generated by artificial intelligence.
Are Any Manufacturers Bundling Traditional Plagiarism Software with AI Detection Tools?
The process of detecting plagiarized content and AI-generated content is distinct. The former requires comparing a submitted work to the massive inventory of work from which the plagiarism software draws. The latter requires an intricate understanding of how generative AI works and how to train an AI to spot the hallmarks of AI-generated writing.
A few organizations have been able to marry traditional plagiarism checkers with cutting-edge AI detection, and they include:
-
Turnitin, which has continued to be a leading name in the academic misconduct space by adapting its offerings to the AI era
-
Originality.ai, whose suite of products includes a “Bulk Scan” that evaluates a given work for both plagiarism and AI-generated content
-
Copyleaks, which offers both a plagiarism checker and an AI detection tool to its users
Universities are among the users of these and other products. While it is understandable that universities would need to check students’ work for signs of plagiarism, schools are not always aware of the flaws that render these tools unreliable.
Are These Tools Reliable?
To answer this question, we must look no further than the words of those who sell them.
Turnitin’s website admits that its technology is prone to raising “false positives.” While Copyleaks says that its false-positive rate is “less than .03%,” it admits that false positives do happen on its platform. Plus, anyone who understands marketing would reasonably suspect that Copyleaks’ figure has been the result of some fair share of massaging.
Each case is different. If a plagiarism detector reveals that 100% of a student’s essay matches a work that has existed on the internet for decades, that plagiarism detector has likely done its job.
On the other hand, an AI detector that accuses a student of misconduct based on a few dubious indicators may be incorrect in its judgments.
The potential for these technologies to make mistakes means that every accusation they generate warrants intense scrutiny.
What Happens If My Work Is Flagged for Plagiarism or AI-Generated Content?
This depends on how your university uses and relies upon these tools. In some cases, an alert from Turnitin or Originality.ai may be enough to trigger a full-on disciplinary investigation. In other cases, such an alert may only be grounds for an informal conversation with the student.
You should proceed as if the worst-case scenario is a possibility. Allow our Student Defense Team to advise and protect you, as some schools regard AI detectors and plagiarism checkers as foolproof. If you do not present convincing evidence of innocence, you could be punished based on a false or exaggerated alarm from one of these unreliable resources.
Don’t wait to protect yourself. Call the LLF National Law Firm Student Defense Team today at 888-535-3686 or contact us online as soon as possible.