November 30, 2023, was a significant date. The world woke up to the news of the launch of ChatGPT, an AI chatbot capable of ‘understanding’ and generating human-like text for the first time. Within the first few weeks of being open to the public, ChatGPT garnered millions of users from all around the world. People began leveraging the power of generative AI technology to generate text summaries, write code, compose poems, solve complex mathematical problems, and much more.
For us in education, the impact of ChatGPT is felt across the spectrum, triggering fear and skepticism and driving many to doubt its utility within educational settings. Concerns about plagiarism, copyright, and privacy were echoed among teachers and educators, leading some school districts to ban ChatGPT in their systems and devices.
However, carceral solutions are not the answer. Generative AI is here to stay, and we need to embrace it. A ‘new normal’ is being created, and we must not replicate the mistakes made during the pandemic when everyone was forced to adapt to new ways of teaching and learning, which were entirely dependent on technology.
That said, the purpose of today’s post is to share a quick overview of the evolution of ChatGPT. Before I do that, let’s clarify the difference between three words that are sometimes used interchangeably but mean different things: OpenAI, GPT, and ChatGPT. OpenAI is the organization responsible for developing ChatGPT. It was founded in 2015 by Elon Musk, Sam Altman, Greg Brockman, Wojciech Zaremba, and Ilya Sutskever, with Sam Altman currently serving as CEO.
GPT, which stands for Generative Pre-trained Transformer, is an AI model developed by OpenAI and is used to train AI models like ChatGPT. ChatGPT is an AI chatbot developed by OpenAI that uses GPT to perform numerous natural language processing tasks, from writing to generating code to composing sonnets.
The Evolution of ChatGPT
Here is a quick overview of the evolution of ChatGPT over the years:
- June 2018: The launch of GPT-1 marked the introduction of one of the early large language models, featuring 117 million parameters. This model set the foundational architecture for the subsequent developments in generative AI.
- February 2019: GPT-2 was introduced, significantly upgrading the model with 1.5 billion parameters. This enhancement improved text generation capabilities, enabling the creation of more coherent and sophisticated text.
- June 2020: GPT-3 was released, constituting a major leap in AI technology with its 175 billion parameters. It was employed to perform various tasks, such as drafting emails, writing articles, composing poetry, and generating code, showcasing its versatile applications.
- November 30, 2022: ChatGPT was launched using GPT-3.5, and it quickly went viral. Its ability to understand and generate human-like text attracted millions of users worldwide, highlighting the practical potential of generative AI.
- February 1, 2023: The introduction of ChatGPT Plus, a premium subscription service, provided users with enhanced features, including reduced downtime and early access to new capabilities.
- March 14, 2023: GPT-4 was integrated into ChatGPT, further enhancing the chatbot’s reliability and creativity. This integration marked another significant advancement in the model’s development.
- May 2024: The release of GPT-4o, the latest and most advanced GPT model, brought additional improvements and capabilities, solidifying the position of OpenAI’s technology at the forefront of AI innovation.
This post is adapted from my upcoming book on ChatGPT for teachers which will soon be available through major publishing platforms.
