news
During an AMA on Thursday, OpenAI CEO Sam Altman and other executives responded to a number of queries from online users.
OpenAI CEO Sam Altman and several other company executives hosted an ask-me anything (AMA) session on Thursday. Users were encouraged to ask questions about the AI company’s products, like ChatGPT, or general enquiries about artificial intelligence (AI) and artificial general intelligence (AGI) during the session, which was hosted on the social networking site Reddit. Altman stated during the meeting that while GPT-5 will not be available this year, the business intends to launch “some very good releases” by the end of 2024.
AMA on Reddit hosted by OpenAI staff
The AMA session was hosted on the ChatGPT subreddit. Altman, OpenAI CPO Kevin Weil, SVP of Research Mark Chen, VP Engineering Srinivas Narayanan, and Chief Scientist Jakub Pachocki were among the executives who took part in the Q&A post, which they referred to as “our Reddit launch.” The official X (formerly known as Twitter) handle of OpenAI also posted about the Reddit AMA.
Answering a question about the timeline for GPT-5 or its equivalent’s release, Altman said, “We have some very good releases coming later this year! However, there is nothing that we will refer to as gpt-5. Many reports have confirmed this, and most predict that OpenAI will release the next flagship model sometime in 2025.
Another user asked about the value that SearchGPT or the ChatGPT Search feature brings, Altman said that he finds it to be a faster and easier way to get to the information. Additionally, he emphasised that complex research will benefit more from the web search functionality. “I also look forward to a future where a custom web page can be rendered dynamically in response to a search query,” he continued.
Weil also answered a number of user enquiries. When asked why Sora was delayed, the OpenAI CPO explained that it was because it took longer to perfect the model, get safety and impersonation right, and scale computation. He did not, however, disclose the launch date.
Weil also pointed out that the company’s ‘o’ series AI models, like GPT-4o and o1-preview, will establish themselves as mainstays and continue to be used even after GPT-5 is released. He also disclosed that the AI could be made to sing by adjusting the ChatGPT Advanced Voice Mode.
One user suggested, “Can we please get a hands-free way to end a voice conversation with ChatGPT? The voice chat must always be manually ended when using the iPhone’s action button shortcut. Weil remarked, “I love this idea,” as if impressed by the concept. Sharing right now with the team!”
Research or SVP at OpenAI Additionally, Mike Chen responded to a significant user query regarding AI hallucinations. He described the reason why hallucinations from AI models persist as a “fundamentally hard problem.” This is due to the fact that AI models learn from human-written text, and human error is frequently incorporated into the large language models’ (LLMs’) core dataset.
“We think RL will also help with hallucinations – when we can programmatically check whether models hallucinate, we can reward it for not doing so,” Chen continued. “Our models are improving at citing, which grounds their answers in trusted sources,” Chen said.