Welcome to 9to5Neural. AI progresses rapidly, and we’re here to help you stay informed. Recently, Elon Musk initiated an attempt to purchase OpenAI, leading Sam Altman to informally decline while jesting about acquiring Twitter. While that was amusing, the AI landscape seems to be stabilizing — HOLD ON, WHAT?! OpenAI just revealed ChatGPT-5, which is set to launch this year?
In a strategic move, OpenAI’s Sam Altman has regained the spotlight by outlining the future of ChatGPT. Altman shared what he termed an OpenAI roadmap update regarding both the GPT-4.5 and GPT-5 AI models.
He begins by recognizing the existing complexity of the ChatGPT experience.
We aim to improve our communication about our planned roadmap, and we must simplify our product offerings. Our goal is for AI to simply “work” for you; we understand that our model and product offerings have become quite convoluted.
The rapid pace of product launches has contributed significantly to the complexity within ChatGPT. For ChatGPT Plus users, the application currently provides the following model options along with descriptions:
- GPT-4o: Ideal for most inquiries
- GPT-4o with scheduled tasks: Instruct ChatGPT to follow up later
- o1: Utilizes advanced reasoning
- o3-mini: Quick at advanced reasoning
- o3-mini-high: Excellent for coding and logic tasks
- GPT-4o mini: Faster response for most questions
- GPT-4: Legacy model
The difficulty for users lies in selecting which type of model to begin with: a large language GPT model or a reasoning o-series model. OpenAI’s challenge is to identify which model will deliver the responses you require. That is what OpenAI is addressing next.
We share your frustration with the model picker and aspire to return to a seamless unified intelligence. Our next release will be GPT-4.5, internally referred to as Orion, marking our last non-chain-of-thought model. Subsequently, one of our primary goals is to integrate o-series and GPT-series models into a single system that can utilize all our tools, know when to engage in extensive reasoning, and be generally applicable across various tasks.
Altman clarified in the announcement that the previously suggested Orion model will be GPT-4.5, arriving within weeks. While it won’t feature the DeepSeek-like chain-of-thought capability found in o3, ChatGPT-5 and future versions will incorporate chain-of-thought functionality to illustrate the model’s problem-solving processes.
ChatGPT-5, Altman shares, will serve as a cohesive model connecting LLM and reasoning models, with chat features being accessible for free. He later mentioned that ChatGPT-5 will be available in mere months.
In both ChatGPT and our API, we are introducing GPT-5 as a system that consolidates much of our technology, including o3. We will no longer offer o3 as a standalone model. The free tier of ChatGPT will include unlimited chat access to GPT-5 at the standard intelligence setting (!!), subject to abuse limits.
Lastly, Altman emphasizes that GPT-5 will provide enhanced intelligence for paid subscribers, potentially referring to more intensive resource capabilities.
Plus subscribers will have the option to operate GPT-5 at a higher intelligence level, while Pro subscribers can utilize GPT-5 at an even greater intelligence capacity. These models will encompass voice, canvas, search, deep research, and additional features.
One concern I have about Altman’s update is that it wasn’t composed entirely in lowercase. Just saying.
In a prior message, Altman also mentioned that OpenAI plans to offer 10 deep research queries each month for the $20/month tier, while free users will start with two deep research queries. Deep research currently necessitates OpenAI’s $200/month ChatGPT Pro plan.
No capitalization for that one.
Stay tuned for more insights into the latest AI advancements in the upcoming edition of 9to5Neural— exclusively on DMN! You can check out the previous issue here.
