OpenAI’s CEO, Sam Altman, recently highlighted the current GPU shortage and raised the question of whether CPUs are being overlooked in generative AI use cases. GPUs are well-suited for deep learning and reinforcement learning applications due to their ability to divide large tasks into smaller ones and execute them in parallel. LLMs, such as GPT-3 and GPT-4, require vast dataset processing at high speeds, which GPUs can provide. OpenAI’s ChatGPT has gained immense popularity due to its ability to produce human-like conversations.
