The high-profile AI startup Anthropic released a new version of its Claude chatbot, saying it outperforms other leading chatbots on a range of standard benchmark tests, including systems from Google and OpenAI.
Dario Amodei, Anthropic’s chief executive and co-founder, said the new technology, called Claude 3 Opus, was particularly useful when analysing scientific data or generating computer code.
Chatbots like ChatGPT can answer questions, write term papers, generate small computer programs and more. They may also generate false or misleading information, much as people do. When OpenAI released a new version of its technology called GPT-4 last spring, it was widely considered the most powerful chatbot technology used by both consumers and businesses. Google recently introduced a comparable technology, Gemini.
But the leading AI companies have been distracted by one controversy after another. They say the computer chips needed to build AI are in short supply. Still, the technology continues to improve at a remarkable pace.Anthropic claims that its Claude 3 Opus technology outperforms both GPT-4 and Gemini in mathematical problem solving, computer coding, general knowledge and other areas.
Claude 3 Opus is open to consumers who pay $20 per month for a subscription. A less powerful version, called Claude 3 Sonnet, is available for free. The firm also rolled out Haiku — another chatbot.
The company allows businesses to build their own chatbots and other services using the Opus and Sonnet technologies.
Both versions of the technology can respond to images as well as text. These can analyse a flowchart, for instance, or solve a math problem that includes diagrams and graphs. But the technology cannot generate images.
More From This Section
Chatbots capable of mimicking human conversation have become an increasing focus — with fast tech advances fuelling an investing frenzy. Although chatbots themselves are by no means new, the technology powering Claude and competitors’ bots is a more powerful tool known as a large language model. But the technology has issues. For example, the chatbots are prone to saying things that aren’t true, an issue sometimes referred to as hallucinations. “These models are still just trained to predict the next word — it’s very, very hard to get to zero percent hallucination rate,” Anthropic President Daniela Amodei said. In its latest launch, the company has tried to address the problem, a priority for Anthropic customers, Amodei said. The company said the new versions of Claude software are twice as likely to offer correct answers to questions.
©2023 The New York Times News Service