Unless you’ve been living in a cave for the past few months, you’ve probably noticed that the Internet is abuzz about ChatGPT. This AI-based chatbot system uses natural language processing (NLP) to generate conversations.
OpenAI, an artificial intelligence research company founded by Sam Altman and Elon Musk, launched ChatGPT on November 30, 2022. The AI chat app was a definite instant hit as it amassed 1 million users less than a week after its release and gained extensive coverage across media.
ChatGPT is a variant of GPT 3, a language prediction model dubbed "the most powerful AI system in the world" by Forbes. The hype over ChatGPT was so intense that it resulted in Google declaring "code red" due to its possible risk to the SEO industry, according to the New York Times.
Despite its flaws, the insurmountable potential of ChatGPT is set to make changes across various language-specializing industries. Kevin Roose, a New York Times writer, stated, "ChatGPT is, quite simply, the best artificial intelligence chatbot ever released to the general public."
$10B Microsoft Investment in the Horizon
The current valuation of ChatGPT is around $29B and is expected to rise, given that Microsoft is considering investing as much as $10 billion in the app. Microsoft plans to integrate ChatGPT in Bing to surpass Google as a search engine. Before this, Microsoft invested $1B in OpenAI.
In addition, ChatGPT is also discussing selling existing shares with Founder Fund and Thrive Capital which is estimated to reach up to $300M. In light of this, Altman projected that ChatGPT would generate $1B in profits by 2024.
The Ethical Dilemma in the Education Sector
Given ChatGPT's effectiveness, educational institutions are alarmed that students will use the app to cheat. Rotman School of Management Associate Professor Kevin Bryan goes so far as to say, "you can no longer give take-home exams/homework."
He tested ChatGPT by having the app answer one of his exam questions and remarked, "the OpenAI chat is frankly better than the average MBA at this point.
I have helped run an AI-based entrepreneurship program for years, written papers on the econ of AI, and follow the field quite closely. Nonetheless, I am *shocked* by how good OpenAI's new chat (https://t.co/6PbtqZUziI) is. E.g., you can no longer give take-home exams/homework.— Kevin Bryan (@Afinetheorem) November 30, 2022
Reactions over this issue are split, with some students defending the use of the app, citing that ChatGPT helps to save time. A Singaporean student asserts, "As we move towards a digitalized society, shouldn’t we rely on digital tools to aid us in our day-to-day tasks?"
Regardless, ChatGPT is still deemed to pose a threat to learning which led New York City Public Schools to outright ban the app's usage.
New York City Deputy Press Secretary Jenna Lyle explains that the rationale of the ban is "due to concerns about negative impacts on student learning, and concerns regarding the safety and accuracy of the content."
Anti-ChatGPT App Evens the Playing Field
To this end, Princeton Computer Science Student Edward Tian created GPTZero, a ChatGPT detection software that combats what he calls "AI plagiarism." He explained that GPTZero works by examining the complexity and randomness of a written output to determine if it's AI-generated.
On January 2, Tian released a demo showing how GPTZero correctly determined that ChatGPT wrote a LinkedIn post while Frame of Reference, an article from The New Yorker, was written by a human. Just like ChatGPT, GPTZero was a success to the point of crashing "due to unexpectedly high web traffic."
ChatGPT is incredibly limited, but good enough at some things to create a misleading impression of greatness.— Sam Altman (@sama) December 11, 2022
it's a mistake to be relying on it for anything important right now. it’s a preview of progress; we have lots of work to do on robustness and truthfulness.
Altman acknowledges the shortcomings of the ChatGPT, stating that the bot should not be relied on. He explains that it's simply a "preview of progress" and that OpenAI will work further on its development, starting with a watermark to prevent plagiarism.