OpenAI's upcoming GPT-5, the next iteration of its large language model, is anticipated to feature enhanced reasoning capabilities,
improved accuracy, and extended support for videos, as CEO Sam Altman shared on the Bill Gates Unconfuse Me podcast.
Altman addressed concerns about unreliable responses and model understanding issues, expressing confidence in addressing these issues with GPT-5.
The model is set to be fully multimodal, supporting speech, images, code, and video.
Although there were initial speculations about GPT-5 potentially achieving superintelligence, current expectations suggest it will be a more advanced version of existing AI models like GPT-4 or Google's Gemini Ultra.
Altman hinted at GPT-5 being part of an ongoing trajectory toward Artificial General Intelligence (AGI), emphasizing continuous improvement in the next 5 to 10 years.
Furthermore, customization will be a significant improvement, allowing users to tailor how the AI responds, acts, and solves problems.
Addressing reliability issues prevalent in current AI models, Altman emphasized that GPT-5 aims to provide more consistent and reliable responses, avoiding hallucinations.
Balancing the benefits of AI advancements with privacy and security considerations will be crucial as GPT-5 and similar models evolve.
Despite the promising advancements, concerns about AI access to personal data and the potential risks of technological development, such as Skynet-like scenarios, remain valid.