San Francisco, California – OpenAI recently announced that access to GPT-4o’s Advanced Voice feature will be delayed until the fall for most users. Despite this, some fortunate individuals, like Reddit user RozziTheCreator, were able to experience a sneak peek of the next-generation voice assistant. RozziTheCreator shared a recording showcasing a new GPT-4o voice narrating a horror story with accompanying sound effects like thunder and footsteps. The clip gained attention after AI writer Sambhav Gupta highlighted it on X.
However, OpenAI clarified that granting RozziTheCreator access was a mistake, and it has since been rectified. The leaked video provided a glimpse into GPT-4o’s capabilities in a more natural setting, featuring sound effects not previously heard.
RozziTheCreator described the experience as unexpected, noting a difference in the voice but encountering some bugs during the brief encounter with the advanced voice. The snippet of the story captured in the recording hinted at the potential of GPT-4o’s storytelling abilities.
OpenAI’s rollout of new features, including GPT-4o’s advanced voice, has faced delays due to security concerns and infrastructure readiness. The accidental access granted to some ChatGPT users has been acknowledged as an error, with OpenAI confirming that the official rollout for Plus users will begin next month.
Despite the delays, the leaked video of GPT-4o’s capabilities has generated excitement for its full potential. The incident serves as another example of GPT-4o pushing boundaries and highlighting its advanced capabilities. The anticipation for wider access to GPT-4o continues to grow, as users eagerly await its official release.