Subscribe To My Newsletter
Finally, after days of speculation, OpenAI today revealed in a formal presentation the new developments the company is working on.
The presentation was led by Mira Murati, the CTO of OpenAI, and concluded with a live presentation by the research leads Mark Chen & Barret Zoph.
Mira started the presentation by announcing the disclosure of three new products and emphasized the importance of OpenAI’s mission, which is to develop products that are freely accessible to everyone. She highlighted that one demonstration of this commitment is the access to ChatGPT without the need to sign up.
Mira closed the introduction with the OpenAI mission:
“To ensure that artificial general intelligence benefits all of humanity.”
Desktop App & Web UI Update
OpenAI has released a new version of ChatGPT with desktop and UI updates. One of the main benefits is to make the ChatGPT desktop version easier, simpler, and seamlessly integrated into your workflow. Another enhancement in the UI is to make it more natural.
The Announcement of GPT-4o
According to Mira, ChatGPT-4o will be accessible to everyone, including free users. The new flagship model can reason across text, voice, vision, and text in real-time. This chat provides GPT-4 level intelligence but is much faster. It is 2x faster, 50% cheaper, and has 5x higher rate limits compared to GPT-4 Turbo. It is also available in API.
In conclusion, when using GPT-4o, ChatGPT Free users will now have access to:
- GPT-4 level intelligence
- Get responses from both the model and the web
- Analyze data and create charts
- Chat about photos you take
- Upload files for assistance summarizing, writing, or analyzing
- Discover and use GPTs and the GPT Store
- Build a more helpful experience with Memory
An Overview Of GPT Store
Before starting the live research demo, Mira spent some time giving an overview of the GPT Store. She explained that users can find custom ChatGPTs for specific use cases in the store, using examples such as university professors creating content for their students or podcasters creating content for their listeners with these GPTs. Additionally, users can now use Vision to upload screenshots, photos, and documents (containing both text and images). They can start a conversation with ChatGPT about all of this content in 50 different languages to build their own GPTs.
GPT-4o Live Research Demo
During the last 15 minutes of the presentation, Mira Murati and the research leads, Mark Chen & Barret Zoph, conducted a live research demo showcasing the voice, vision, desktop code capabilities, and real-time translation.
Click here for more insights and declarations from Sam Altman, CEO Of OpenAI
