OpenAI, the organisation behind the popular ChatGPT, has embarked on a new journey with its “Next Frontier Model.” While it’s unlikely to be called GPT-5, this model represents a significant leap forward in AI research and development. This new model is expected to bring the next level of capabilities in the company’s pursuit of artificial general intelligence (AGI).
The Journey So Far
The first version of ChatGPT, based on GPT-4, made waves when it debuted in November 2022. Its generative capabilities captured public attention, leading to widespread adoption. Since then, OpenAI has continuously improved the underlying model, addressing limitations and enhancing performance.
The Next Frontier Model
The term “frontier model” is used in the AI industry to describe a new AI system designed to push the boundaries of current capabilities. AGI, on the other hand, refers to a hypothetical AI system with human-level abilities to perform novel, general tasks beyond its training data.
This is unlike narrow AI, which is trained for specific tasks. OpenAI anticipates that the resulting system from training this next frontier model will unlock newer AI capabilities, bringing us closer to AGI. However, whether this new frontier model is intended to be GPT-5 or a step beyond that is currently unknown.
Safety and Security Committee
In view of the training of this new model, OpenAI has announced the formation of a new “Safety and Security Committee” to oversee risk management for its projects and operations. This committee, led by OpenAI directors Bret Taylor (chair), Adam D’Angelo, Nicole Seligman, and Sam Altman (CEO), will be responsible for making recommendations about AI safety to the full company board of directors.
The committee’s first task will be to evaluate and further develop the company’s processes and safeguards over the next 90 days. At the end of this period, the committee will share its recommendations with the full board, and OpenAI will publicly share an update on adopted recommendations.
Key Features of the Next Frontier Model
Unprecedented Scale and Capabilities:
The Next Frontier Model pushes the boundaries of large-scale language models. It surpasses existing models in terms of size, training data, and capabilities. Expect the Next Frontier Model to handle a wider range of tasks, including complex reasoning, creative writing, and specialised domains.
Safety and Responsible Development:
OpenAI is committed to responsible AI development. The Next Frontier Model undergoes rigorous safety evaluations to minimise risks. The formation of the Frontier Model Forum, in collaboration with Anthropic, Google, and Microsoft, emphasises safety research and best practices.
Path to AGI:
OpenAI explicitly states that the Next Frontier Model aims to bring us closer to Artificial General Intelligence (AGI). While AGI remains a long-term goal, the Next Frontier Model represents a significant step in that direction.
Differences from ChatGPT
Model Size and Complexity:
The Next Frontier Model is larger and more complex than ChatGPT. It can handle more diverse tasks and exhibit improved performance. Expect better context understanding, nuanced responses, and fewer instances of nonsensical outputs.
Safety Measures:
The Next Frontier Model undergoes thorough safety checks. Unlike ChatGPT, which occasionally produced harmful or biased content, the Next Frontier Model aims for higher safety standards.
Industry Collaboration:
The Frontier Model Forum brings together industry leaders to share knowledge and promote responsible AI development. ChatGPT lacked such collaborative efforts, but the Next Frontier Model benefits from collective expertise.
OpenAI’s Next Frontier Model represents a significant leap forward in AI research. As it continues to evolve, we can expect more powerful, safer, and versatile language models. While AGI remains the ultimate goal, the Next Frontier Model paves the way for exciting advancements in natural language understanding and generation.
ALSO READ: Tensions rise amongst countries developing Artificial Intelligence
