ChatGPT is set to become significantly more effective.
OpenAI announced on Monday a new version of its artificial intelligence language model, GPT-4o, which enhances ChatGPT’s intelligence and usability. This model is an upgrade over the previous GPT-4 model, introduced just over a year ago, and will be accessible to all users at no cost.
During the company’s demonstration, it was shown that GPT-4o will transform ChatGPT into a virtual personal assistant capable of conducting spoken conversations in real-time. It will also support interactions through text and visual inputs, allowing it to analyze and discuss images, documents, and other visual data uploaded by users.
Mira Murati, OpenAI’s Chief Technology Officer, highlighted that the new ChatGPT will include memory capabilities to retain information from past interactions and will feature real-time translation services.
“This represents a significant advancement in user-friendliness,” Murati stated during the demo at OpenAI’s headquarters in San Francisco. “Interactions will become much more natural and much easier.”
This release aims to position OpenAI ahead of competitors like Google and Meta, who are also developing advanced language models. Google is expected to unveil updates to its Gemini AI model, which, like GPT-4o, is multimodal, at its upcoming I/O developer conference.
The GPT-4o update also anticipates upcoming AI announcements from Apple at its Worldwide Developers Conference, which may introduce new AI integrations into iPhones or iOS.
The benefits of this new GPT model extend to Microsoft as well, which has heavily invested in OpenAI and incorporates its AI technology into its products.
During the demonstration, OpenAI executives showcased ChatGPT’s ability to converse naturally, solve math problems, narrate bedtime stories, provide coding advice, and even respond to emotional cues. The model was capable of speaking in both a human-like and a robotic voice, and it supported conversations in over 50 languages.
CEO Sam Altman remarked on the new model’s capabilities, noting, “The new voice (and video) mode is the best computer interface I’ve ever used. It feels like AI from the movies.”
Murati also announced that a desktop app for ChatGPT with GPT-4o capabilities will soon be launched, expanding the ways users can engage with the technology. Additionally, the GPT-4o model will be available for developers through OpenAI’s GPT store, even for those without a subscription.
The updated features are scheduled to roll out to ChatGPT in the following months. While free users will have limited interactions with the new model, paid subscribers will enjoy extended access.
With over 100 million existing users, these enhancements are poised to attract even more people to use OpenAI’s technology, especially as competitors like Google and Meta integrate AI into widely-used platforms such as Instagram and Google Assistant.
