OpenAI has launched its ChatGPT Pro plan, a premium subscription service that grants access to the company's souped-up o1 pro mode, touted to provide more reliable and comprehensive answers to complex questions. The catch? It comes with a hefty price tag of $200 per month, leaving many in the AI community wondering if it's worth the cost.
The o1 pro mode is an enhanced version of OpenAI's o1 reasoning model, designed to excel in answering questions related to science, math, and coding. However, early adopters have been quick to test the model's limits, with some users asking it to create whimsical designs, such as unicorns and "crab-based" computers, while others have probed its philosophical capabilities, like waxing poetic on the meaning of life.
Despite the initial excitement, many experts and users have expressed skepticism about the value proposition of o1 pro mode. British computer scientist Simon Willison questioned the lack of concrete examples demonstrating the model's superiority over the standard o1, citing the need for tangible proof to justify the substantial price increase. Others have pointed out that o1 pro mode struggles with tasks like Sudoku and optical illusion jokes, which are easily solvable by humans.
OpenAI's internal benchmarks show that o1 pro mode performs only slightly better than the standard o1 on coding and math problems, leading some to wonder if the premium price is justified. Even OpenAI CEO Sam Altman has clarified that ChatGPT Pro is not intended for the general public, suggesting that most users will be satisfied with the standard o1 or the Plus tier.
So, who is ChatGPT Pro really for? Experts like Ameet Talwalkar, an associate professor of machine learning at Carnegie Mellon, believe that the high price point may be a significant risk, and it remains to be seen if there is a strong appetite for this functionality. Others, like UCLA computer scientist Guy Van den Broeck, have questioned the price point, wondering if pricey reasoning models will become the norm.
One possible explanation for the high cost is that o1 pro mode is intended for specialized fields, such as healthcare, where accuracy and reliability are paramount. OpenAI plans to grant free access to ChatGPT Pro to a select group of medical researchers at leading institutions, which could unlock significant value in critical applications.
However, as Bob McGrew, OpenAI's former chief research officer, noted, the limitations of a simple, text-based interface may hinder users from fully leveraging o1 pro mode's capabilities, a phenomenon he calls "intelligence overhang." This raises questions about the long-term viability of ChatGPT Pro and whether it can justify its hefty price tag.
As the AI community continues to grapple with the implications of ChatGPT Pro, one thing is clear: OpenAI's premium offering has sparked a lively debate about the value of advanced AI capabilities and the importance of transparency in showcasing their benefits. Whether ChatGPT Pro will find its niche in the market remains to be seen, but one thing is certain – the conversation has only just begun.