We are minutes away from OpenAI’s Spring Update livestream.
OpenAI’s co-founder and CEO, Sam Altman, revealed on X/Twitter this past Friday that the company, a pioneer in generative AI technology, is poised to unveil significant updates to its flagship chatbot, ChatGPT, and the GPT-4 large language model.
Reports and rumors had been circulating for weeks, but on Friday, Altman quashed these speculations with a post on X stating: “Reports and rumors had been circulating for weeks, but on Friday, Altman quashed these speculations with a post on X stating:
“not gpt-5, not a search engine, but we’ve been hard at work on some new stuff we think people will love! feels like magic to me.”
Welcome GPT-4o
GPT-4o is now introduced and Mira Murati, the CTO of OpenAI defined it as its “newest flagship model.” GPT-4o is able to reason “across voice, text and vision.”
GPT-4o is introduced here here’s 5 very cool use cases
“An important part of our mission is being able to make our advanced AI tools available to everyone for free,” including removing the need to sign up for ChatGPT. During a demonstration, they unveiled a Mac desktop application that includes the Voice mode, previously only accessible on mobile devices. This feature is not a voice assistant as was speculated.
Murati described GPT-4o as altering the traditional ways in which humans and machines collaborate. “When we interact with one another there is a lot we take for granted,” she said. She noted that GPT-4o has the capability to process information across voice, text, and vision modalities. “This allows us to bring the GPT-4-class intelligence to our free users,” she added, mentioning that this has been in the pipeline for several months.
ChatGPT is used regularly by over 100 million individuals, and GPT-4o is notably more resource-efficient than its predecessors. As a result, they are now able to offer GPTs (customizable chatbots) in the free version of ChatGPT. Users will soon have the ability to employ tools for data, coding, and vision, facilitating image analysis without the need for a paid subscription. These updates will be implemented in the coming weeks.
The substantial enhancements to the free version of ChatGPT, which now includes features that were once exclusive to the ChatGPT Plus subscription, prompt questions about the value of the $20 monthly fee.
OpenAI’s CTO, Mira Murati, clarifies that the primary advantage for paid subscribers is the ability to make five times as many daily requests to GPT-4o compared to those on the free plan.
You can now voice chat ChatGPT in real-time!
A significant enhancement in GPT-4o is its live speech capability. This model operates in a speech-to-speech format, directly processing audio inputs without first transcribing them into text.
During the demonstration of this feature, an OpenAI staff member simulated heavy breathing into the voice assistant, which successfully provided suggestions for improving breathing techniques, enhancing singing and robotic voice qualities, and offered guidance to uplift spirits.
In addition, GPT-4o introduces a groundbreaking feature that enables users to livestream receive real-time assistance from ChatGPT. This capability exemplifies how the platform can interactively aid in problem-solving, showcasing its potential as an educational tool.
For example, students or researchers can present complex equations during a livestream, and ChatGPT can immediately provide explanations and solutions, enhancing learning and comprehension in an innovative way.
ChatGPT Desktop app
With a remarkably natural voice interface on a Mac, ChatGPT demonstrated its ability to observe code being written in real-time and conduct an analysis. It could also articulate its observations, including identifying any potential problems.
The vision functionalities of the ChatGPT Desktop application appear to extend to desktop viewing. In the demonstration, it was capable of examining a graph and offering genuine feedback and insights.
ChatGPT is now a real-time translator
In a demonstration, the OpenAI team showcased the capability of ChatGPT Voice as a live translation tool. It effectively translated sentences from Italian spoken by Mira Murati into English, and then converted responses back from English into Italian.
