OpenAI Launches Multimodal GPT-4 as Anthropic Unveils Safer Claude Chatbot
OpenAI releases GPT-4, a highly capable language model that processes both text and images, while rival Anthropic debuts its Claude chatbot focused on safety. Both AI systems launch on the same day, marking a significant leap in artificial intelligence capabilities.
OpenAI introduces GPT-4, a powerful new large language model that answers complex questions more accurately than its predecessor. Unlike previous versions, this new model accepts both text and images as input, allowing users to upload photos and ask questions about them. Microsoft reveals that its Bing chatbot already runs on a customized version of this advanced system.
Rival Anthropic launches its own chatbot named Claude on the exact same day. Backed by recent funding from Google, Anthropic designs Claude to be highly user-friendly and significantly less likely to generate harmful responses. Both Claude and GPT-4 focus heavily on text processing tasks, though they approach the market with different safety philosophies.
GPT-4 demonstrates exceptional reasoning skills by scoring in the top 10% on a simulated bar exam and reaching the top 1% on an International Biology Olympiad test. OpenAI develops the model over two years using a dedicated supercomputer built in partnership with Microsoft Azure. The company notes that GPT-4 surpasses GPT-3.5 in reliability and creativity when handling highly nuanced instructions.