ChatGPT has a more Manageable and Secure Competitor

Anthropic developers have introduced a new large language model Claude, which is a more humane and manageable analogue of ChatGPT. According to the developers, their neural network produces harmful results much less often, it is easier to communicate with it and easier to manage compared to any other AI-based chatbots.

Anthropic has released two versions of its AI: the high performance and modern Claude and the lighter, cheaper and faster Claude Instant. Unlike ChatGPT, Claude is not available through the public interface. It is impossible to communicate with her on the Anthropic website – only through the API. Therefore, other developers can integrate Claude into their products.

As Anthropic acknowledges, Claude still has its flaws. So, the AI ​​can still spew nonsense. For example, during the testing phase, he mentioned chemicals that do not exist in nature (or does the AI ​​​​know something?) Also, Claude is not as good at programming and mathematics, unlike ChatGPT.

Claude has already been tested in DuckAssist by DuckDuckGo, part of Notion AI and an AI chat app called Poe built by Quora. Developers recently raised $300 million in investment, bringing the company’s total valuation to $4.1 billion.