San Francisco-based Anthropic introduces language model Claude as OpenAI competitor
Anthropic, an AI company backed by Alphabet Inc, announced on Tuesday the launch of a new large language model named Claude, which directly competes with Microsoft Corp-backed OpenAI's ChatGPT. Large language models are algorithms trained to generate text by feeding them human-written training data. In recent years, researchers have improved the performance of these models by significantly increasing the amount of data and computing power used to train them.
Claude is built to perform similar tasks to ChatGPT, such as writing computer code and editing legal contracts. However, Anthropic has put a greater emphasis on producing AI systems that are less likely to generate harmful or offensive content, such as instructions for computer hacking or making weapons. This concern for AI safety was highlighted last month when Microsoft limited queries to its new chat-powered Bing search engine after the chatbot produced unsettling responses during an extended conversation.
Anthropic's approach to AI safety is different. Claude is trained with a set of principles, and rather than avoiding dangerous topics, it is designed to explain its objections based on its principles. Richard Robinson, CEO of Robin AI, a London-based startup that uses AI to analyze legal contracts, praised Anthropic's approach, stating that Claude was better at understanding dense legal language and less likely to generate strange responses than OpenAI's technology.
Overall, Anthropics focus on AI safety sets it apart from competitors and may attract businesses that require AI systems with high levels of ethical standards.
Claude is built to perform similar tasks to ChatGPT, such as writing computer code and editing legal contracts. However, Anthropic has put a greater emphasis on producing AI systems that are less likely to generate harmful or offensive content, such as instructions for computer hacking or making weapons. This concern for AI safety was highlighted last month when Microsoft limited queries to its new chat-powered Bing search engine after the chatbot produced unsettling responses during an extended conversation.
Anthropic's approach to AI safety is different. Claude is trained with a set of principles, and rather than avoiding dangerous topics, it is designed to explain its objections based on its principles. Richard Robinson, CEO of Robin AI, a London-based startup that uses AI to analyze legal contracts, praised Anthropic's approach, stating that Claude was better at understanding dense legal language and less likely to generate strange responses than OpenAI's technology.
Overall, Anthropics focus on AI safety sets it apart from competitors and may attract businesses that require AI systems with high levels of ethical standards.
