OpenAI CEO confirms no plans for GPT-5 development amidst AI safety concerns

OpenAI CEO Sam Altman has confirmed that the company is not currently training GPT-5, the presumed successor to its AI language model GPT-4. This comes amidst concerns about the safety of future AI systems expressed in an open letter, which Altman criticized as "missing most technical nuance about where we need the pause." However, while OpenAI might not be working on GPT-5, it is still expanding the potential of GPT-4 and considering safety implications of such work. The fallacy of version numbers, which assumes that numbered tech updates reflect definite and linear improvements in capability, hinders the debate about AI safety. The focus should be on capabilities and demonstrations of what these systems can and can't do rather than just assigning a number to a concept of intelligence. Therefore, Altman's confirmation won't alleviate concerns about AI safety.