OpenAI to Introduce AI Assistant with Sarcasm Detection, Multimodal Features
OpenAI is working on developing an AI voice assistant with visual and audio capabilities, with sarcasm detection, designed to enhance automated customer service, as reported by The Information.
This new technology will not only serve as a tool that engages with users through the audio, but also will be able to recognize objects and images, among other features, according to sources familiar with the matter.
The report also mentioned that these features will include a better understanding of image and audio, as well as improved logical reasoning.
The new AI assistant is expected to be unveiled during OpenAI’s scheduled livestream today on the upcoming 13th of May, to announce some new updates related to their GPT technology.
It could provide a range of functions, such as acting as a tutor for students who are working on papers or solving math problems and helps users to understand information from their surroundings, such as translating signs and giving them guidance when required, like providing them with instructions on how to fix something if ever they requested.
It is worth mentioning that despite these major advancements, the new AI multimodal will surely face some challenges. Like any other AI system, this AI assistant can experience “hallucinations” by generating responses that lack verified evidence.
The AI company’s latest release could potentially be integrated into the company’s famous chatbot, ChatGPT, to enhance its capabilities, enabling it to generate more accurate and detailed answers across a wider range of requests.
As for comedians and customers, they should be aware that this new tool will be able detect their insults and sarcasm.
Updated on Tuesday, May 14
During yesterday’s event, on Tuesday May 13, OpenAI launched GPT-4o, an updated version of the GPT-4.
In addition to the above-mentioned upcoming features, this new version can do real-time interactions actions with users. For example, while the chatbot is generating a voice response, it could be interrupted by users in case they want to add or remove some information.
It is also worth mentioning that this new version will be available for free for both paid and free users, with the paid one enjoying greater limits. However, both users will have access to “browse” feature that allows ChatGPT to generate up-to-date information from the web.
Inside Telecom provides you with an extensive list of content covering all aspects of the tech industry. Keep an eye on our Intelligent Tech sections to stay informed and up-to-date with our daily articles.