Anthropic has expanded the reach of its AI chatbot, Claude, making it accessible to more users in the US and UK. The company touts Claude as easier to converse with and describes it as a friendly personal assistant that can perform a range of tasks using natural language.
Claude's capabilities include creating summaries, writing code, and translating text, much like Google's Bard or Microsoft's Bing chatbot. However, Anthropic emphasizes Claude's distinct conversational tone, complete with a sense of humor and a guiding set of principles. These features enable it to revise its responses independently, negating the need for human moderation.
Introducing Claude 2! Our latest model has improved performance in coding, math and reasoning. It can produce longer responses, and is available in a new public-facing beta website at https://t.co/uLbS2JNczH in the US and UK. pic.twitter.com/jSkvbXnqLd
— Anthropic (@AnthropicAI) July 11, 2023
The latest version, Claude 2, brings several improvements over the first version. It can craft longer responses, showcases enhanced mathematical, coding, and reasoning skills, and is better at delivering harmless responses, reducing the chances of harmful content. For example, Claude 2 scored higher on a bar exam's multiple choice section than its predecessor.
However, Claude 2, unlike Bard and Bing, is not connected to the internet and is trained on data up until December 2022, which means it can't provide up-to-the-minute information on current events. Users can still feed Claude with recently published websites or webpages, and the bot should be capable of answering questions about it.
Anthropic has expanded Claude's context window to approximately 75,000 words, allowing users to upload extensive text for the bot to parse. This is significantly more than the roughly 3,000-word limit of models like ChatGPT, enabling Claude to handle tasks such as summarizing extensive research papers.
