Anthropic, a prominent AI startup, has unveiled an upgraded version of its Claude chatbot, named Claude 3 Opus. According to Dario Amodei, Anthropic’s CEO and co-founder, the new technology surpasses leading chatbots, including those from Google and OpenAI, across various standard benchmark tests. Claude 3 Opus exhibits exceptional performance, particularly in tasks involving scientific data analysis and computer code generation.
Chatbots like ChatGPT, capable of answering questions, crafting term papers, and generating computer programs, have become influential tools. Despite the release of powerful technologies like OpenAI’s GPT-4 and Google’s Gemini, the AI industry has faced controversies and challenges related to the scarcity of computer chips. Nevertheless, technology advancements persist at an impressive rate.
Anthropic asserts that Claude 3 Opus excels in mathematical problem-solving, computer coding, general knowledge, and other domains, outperforming GPT-4 and Gemini. The subscription-based service is available to consumers for $20 per month, while a free version, Claude 3 Sonnet, offers less powerful capabilities. Additionally, Anthropic introduced Haiku, another chatbot, and provides businesses with the opportunity to build their own chatbots using the Opus and Sonnet technologies.
Both versions of the technology demonstrate the capability to respond to both text and images, enabling tasks such as analyzing flowcharts and solving math problems with diagrams and graphs. However, they do not generate images. While chatbots with human-like conversation abilities have become a focal point of technological development, challenges persist, such as the propensity for these models to produce inaccurate information, known as hallucinations. Anthropic President Daniela Amodei acknowledges this issue and highlights the company’s efforts in its latest launch to reduce the hallucination rate, claiming that the new versions of Claude software are twice as likely to provide correct answers to questions.
