OpenAI kicked off its “12 Days of OpenAI” event by unveiling the fully functional version of its o1 reasoning model and a new premium subscription tier, ChatGPT Pro. However, this unlimited access comes at a steep price of $200 per month.
The o1 model, previously known as Project Strawberry, was initially previewed in September alongside the lighter o1-mini model for ChatGPT Plus subscribers. Unlike standard Large Language Models (LLMs), o1 is a reasoning model capable of self-fact-checking before generating responses, reducing the likelihood of hallucinations, albeit at the cost of slower response times.
Despite the slower speed, o1 can deduce answers to complex queries in technical fields like science, coding, and mathematics faster than humans. OpenAI’s internal testing shows the new o1 version reduces “major errors” on complex real-world questions by 34% compared to its preview version.
glasses and chatgptImage: A person wearing glasses is looking at a computer screen displaying a ChatGPT interface. This visually represents the interaction between users and the AI chatbot.
ChatGPT Plus and Team users can access the new model through the ChatGPT website’s model selection tool. Enterprise and Edu subscribers will gain access next week. While initial file uploads and web access are unavailable with the new o1, OpenAI plans to add these features in the coming months.
The new ChatGPT Pro subscription, priced at $200/month, offers unlimited access to all OpenAI models, including GPT-4o, the full o1, and Advanced Voice Mode. Pro subscribers also gain exclusive access to o1 Pro mode, which leverages increased computing power for more accurate and comprehensive answers to complex questions.
ChatGPT Pro: Enhanced Accuracy and Comprehensive Responses
OpenAI informed TechCrunch that o1 Pro mode delivers more reliable, accurate, and comprehensive responses, particularly in areas like data science, programming, and case law analysis. Compared to both o1 and its preview version, o1 Pro mode performs significantly better on challenging machine learning benchmarks across math, science, and coding. Notably, a 75% reduction in errors was observed for simpler coding competition questions, reflecting typical programming queries.
OpenAI’s Push for Profitability
Facing pressure from investors to curb its significant operational costs, OpenAI has been exploring various revenue-generating strategies. Previous estimates suggested ChatGPT’s daily operating costs reached $700,000. The company has considered incorporating advertising within chat results and proposed charging Enterprise customers upwards of $2,000 per month for an elite subscription tier. The New York Times also reported OpenAI might increase Plus-tier subscription fees to $44 per month by 2029.
The Future of ChatGPT Pricing
The question remains whether users will embrace these higher prices, especially given ChatGPT’s occasional tendency to produce incorrect answers with unwarranted confidence. The high cost of ChatGPT Pro raises questions about its accessibility and the balance between advanced features and affordability.
In conclusion, OpenAI’s new o1 model and ChatGPT Pro subscription offer significant advancements in AI capabilities. However, the hefty price tag associated with Pro raises questions about accessibility and market acceptance. The future will determine whether the value proposition justifies the cost for a broader user base.
