1. OpenAI unveiled GPT-40 model with live voice functionality at a technology conference in Paris
2. ChatGPT Voice can watch and interact with users in real-time, including providing detailed directions on maps
3. OpenAI is focused on entering “product mode” and developing new features like Sora video and Voice Engine, with plans for safe release in the future.
OpenAI introduced its GPT-4o model with live voice functionality at an event earlier this month, generating significant excitement. During a demonstration at a technology conference in Paris, OpenAI showcased how the AI could engage with an audience in French. The new capabilities will revolutionize technology interactions once they undergo security testing before widespread availability.
One standout feature of the ChatGPT Voice functionality was its ability to process visual information. The AI accurately identified landmarks from a sketch and provided detailed directions from the current location. The demonstration also showcased how ChatGPT Voice could assist with coding in real-time, suggesting improvements and modifications as code was written.
OpenAI is shifting towards product development and productivity tools, although it remains focused on advancing artificial general intelligence. The Sora video creation tool and Voice Engine, capable of generating deepfakes and multilingual voice cloning, were previewed during the event. These tools are still in development to ensure responsible release amid concerns about disinformation potential.
Overall, the advancements presented by OpenAI, such as live voice interaction, visual processing, and real-time coding assistance, indicate a significant leap in AI capabilities. The potential for transformative technologies like Sora and Voice Engine highlights the need for responsible deployment to mitigate misuse for creating misleading content.