AI now captures what you film in real time

ChatGPT’s advanced voice mode takes a significant step in the evolution of artificial intelligence with the integration of real-time video. This advance, announced by OpenAI on December 12, now allows ChatGPT to “see” its user using the smartphone camera, which transforms interactions with the chatbot.

With this new feature, conversations become much more interactive and immersive, providing an additional dimension to the user experience. Currently being deployed, this update promises to make exchanges with ChatGPT more natural and dynamic, by allowing more personalized communication.

Thus, the AI ​​no longer just sees, it can also remember what it observes. For example, she may remember the name of an employee wearing a striped accessory or that of another person wearing a yellow cap. With this new feature, OpenAI seeks to “enrich and make conversations more useful,” by allowing ChatGPT to share “real-time visual context.” This advancement offers users a more immersive experience, where AI can, for example, guide them through the different steps to prepare Moroccan or Chinese tea.

In addition, OpenAI also introduces the possibility of sharing your screen with advanced voice mode. This will allow users to ask for help responding to a message or performing various visual tasks. These new functions will be available to subscribers to ChatGPT Plus (22.99 euros per month), ChatGPT Team (from 25 dollars per user per month) and ChatGPT Pro (200 euros per month).

OpenAI ensured that these features will be offered as soon as possible, without however providing a detailed explanation for this unavailability. This is likely to be related to European regulations, a factor that currently prevents the company from offering its video generator, Sora, in that region.