Many are nervous about the arrival of ChatGPT, the AI chatbot that can write about anything. ChatGPT can write essays, solve math problems, make original jokes, and even write code.
But it doesn’t write like a typical AI tool, which is repetitive and awkward. ChatGPT writes just like a human, and it writes well. VERY well.
In fact, ChatGPT recently passed some pretty tough exams:
- MBA exams at Wharton
- Several law school exams
- Three-part United States Medical Licensing Exam
- Clinical reasoning final at Stanford Medical School
Plagiarism tools can’t really catch ChatGPT. At least not yet. That’s because ChatGPT was trained with about 300 billion words from internet resources, like research, articles, websites, and more. But it doesn’t get its information from the internet.
Sounds confusing, but here’s an analogy for how ChatGPT works:
During your education, you learned from different textbooks, articles, and instructors. But you don’t just regurgitate that information. You understand it, and you can make complex connections to use in different ways.
That’s basically how ChatGPT works. It doesn’t just copy different pieces of content from the internet. ChatGPT really “comes to life” because it receives feedback and review from experts, which refines its responses to sound more human. All of this is why plagiarism tools won’t catch it anytime soon.
To show you how it works, we’ll use a realistic scenario where an instructor wants participants in class to complete a 500 word essay by 11:59 p.m. on Friday.
Instructors can create rules that specify a certain number of hours participants have to complete the written assignment once they begin. Honorlock’s proctoring software and services are available 24/7/365, so they can write their essays anytime they’re ready within your set time frame, and you’ll know it’s being proctored.
With Honorlock’s BrowserGuard, instructors could do the following:
- Prevent access to all other websites, browsers, and applications.
- Allow access to specific websites and applications while blocking all others, including ChatGPT.
This is important because you may want to give participants access to certain internet articles or lectures in PowerPoint as a supplement while they’re writing their essay, but you want to make sure that they can’t access ChatGPT or other AI writing tools. With Honorlock, you can do just that.
Honorlock’s BrowserGuard also records the participants’ desktops and prevents keyboard shortcuts, like copying and pasting. If copy/paste is attempted, the participant receives a notification that it isn’t allowed and the AI marks it as a medium flag.
If you’re only using browser lockdown software, how would you know if:
- The test taker is using a cell phone to access ChatGPT?
- Voice assistants and voice dictation, like Siri and Alexa, are being used?
- A friend is off camera reading ChatGPT responses to the test taker?
- Or if someone else is writing the essay entirely
If your online proctoring platform doesn’t have smart voice detection, a participant could use a voice assistant, like Siri or Alexa, to:
- Open the ChatGPT website
- Use voice dictation to add a specific prompt
- Read the response aloud
Honorlock’s Smart Voice Detection listens for specific keywords or phrases, like “Hey Siri” or “OK Google,” and it records and transcribes audio while the exam is in progress. If it detects potentially dishonest behavior, a live proctor is alerted to review the situation and intervene if necessary.
Honorlock’s online proctoring platform makes it quick to check the room for unauthorized resources, like cell phones, notes, and even other people. And our proctoring software monitors and records behavior throughout the exam as well.
This way, you’ll know that test takers aren’t accessing ChatGPT from other devices or asking friends in the room to do it for them while they’re writing their essay.
Honorlock upholds academic integrity with online exam proctoring that’s monitored by AI and reviewed by humans. Our AI proctoring software monitors exams for potentially dishonest behavior and alerts a live proctor to review (and intervene if necessary) if it detects any issues. Our blended proctoring solution delivers a less intimidating and non-invasive experience for participants.
Honorlock’s online proctoring software can also detect when a test taker attempts to access test bank content using a cell phone or other devices during the proctored assessment activity.
Honorlock’s Search and Destroy™ technology scours the internet automatically for leaked test content and gives instructors the ability to send takedown requests with one click. This way, you’ll know your test content isn’t available on sites like Quizlet, Chegg, Reddit, and Quora.
You need to know that the person taking the proctored exam is the same person getting credit in the class. Honorlock’s ID verification takes about 60 seconds. It captures a picture of the test-taker along with their photo ID.
ChatGPT is a powerful AI tool, but there are many other tools and different ways for test takers to gain an unfair advantage. Stay ahead of the curve with Honorlock’s blended proctoring solution.
Source:
chat.openai.com/
businessinsider.com/list-here-are-the-exams-chatgpt-has-passed-so-far-2023-1#us-medical-licensing-exam-2