Last month, OpenAI showcased the new features of ChatGPT in a demo that generated excitement among users. The demo highlighted ChatGPT’s ability to speak in synthetic voices and introduced GPT-4o, a more advanced AI system that can respond to tone of voice and facial expressions, as well as engage in more complex conversations.
However, due to the need for further security testing, OpenAI has decided to delay the launch of ChatGPT’s voice assistant features by a month. Initially scheduled for release to some paying subscribers at the end of June, the company is now pushing back the launch until July. The exact timing will depend on meeting security and reliability standards.
OpenAI emphasized the importance of improving ChatGPT’s voice feature to block inappropriate content and create a more authentic conversational experience. Designing AI tools that interpret and imitate emotions introduces new complexities and potential errors into interactions. The company is working to enhance the chatbot’s ability to understand and respond with emotions and nonverbal cues, approaching a more human-like conversational experience with responsibility and caution.
The decision by OpenAI comes at a time when technology companies are striving to advance generative AI. Projects like OpenAI’s ChatGPT have faced unexpected challenges in this competitive landscape. Companies like Google and Microsoft have also encountered hurdles in their AI initiatives, such as generating bizarre responses or spreading biased content. However, OpenAI is determined to address these challenges head-on while ensuring that its technology remains safe for all users.
