OpenAI has announced significant changes to its product lineup, acknowledging customer confusion over its multiple AI models and features. CEO Sam Altman revealed plans to streamline the company’s offerings and integrate various technologies into a unified system.
Current Challenges and Planned Changes
The company currently offers multiple models and subscription tiers, including GPT-4o, o1, o3, and various ChatGPT versions, which has led to confusion among users. Altman specifically criticized the model picker in ChatGPT, where users must select from different models depending on their subscription level.
To address these issues, OpenAI will:
- Release GPT-4.5 (codenamed Orion) as its final non-chain-of-thought model
- Integrate the o-series models into the upcoming GPT-5
- Cancel the standalone release of the o3 model
- Provide unified access based on subscription tiers
New Access Structure
The upcoming GPT-5 will be available across different subscription levels with varying capabilities:
- Free tier users will receive unlimited access at a “standard intelligence setting”
- Plus subscribers will access higher intelligence levels
- Pro subscribers will get the highest level of intelligence
The integrated system will include features such as voice, canvas, search, and deep research capabilities. Notably, the Deep Research feature, previously limited to high-tier subscribers, will be available to free users (2 uses per month) and ChatGPT Plus users (10 uses per month).
These changes represent a significant shift in OpenAI’s product strategy, focusing on user accessibility while maintaining tiered service levels. The company aims to make AI technology “just work” for users without requiring detailed knowledge of different models and their capabilities.
Sources: VentureBeat, TechCrunch, VentureBeat