Anthropic, a prominent AI developer, has implemented noteworthy changes in its commercial terms of service effective January. The update assures increased protection for commercial customers and emphasizes their ownership of all outputs derived from using Anthropic's AI models.
In a commitment to data privacy and intellectual property concerns, Anthropic explicitly pledges not to train its AI models on content from customers of its paid services. The revised terms underscore that the developer does not anticipate obtaining any rights in customer content.
In response to the evolving landscape of AI legal considerations, Anthropic has taken a stance similar to industry peers like OpenAI, Microsoft, and Google. These companies, in the latter half of 2023, introduced policies to defend customers facing legal issues, particularly regarding copyright infringement claims resulting from the use of their technologies.
Anthropic's commitment extends to defending customers against any copyright infringement claims arising from authorized use of its services or outputs. The company also goes further by stating it will cover the costs of approved settlements or judgments resulting from infringements generated by its AI.
The changes aim to provide customers with increased protection and peace of mind, creating a more streamlined experience with Anthropic's AI models, particularly through its API. Anthropic emphasizes that customers using its Claude API and Bedrock, Amazon's generative AI development suite, are covered by these updated terms. Amazon, a major investor in Anthropic, is also the company's primary cloud provider with a substantial $4 billion investment.
Alongside the commercial terms update, Anthropic has introduced improvements to its API, offering users the ability to identify errors in prompts. This change, implemented in the Messages API, aims to enhance prompt construction and identify errors early in the development process.
Anthropic envisions further enhancements to its API, with plans to launch a more robust function calling option in the near future. The company emphasizes making the Claude API more accessible to developers and enterprises, facilitating broader utilization of its AI solutions.
In November of the previous year, Anthropic had already demonstrated its commitment to continuous improvement by releasing the 2.1 update for the Claude model. This update enhanced the model's capabilities in analyzing longer documents while reducing instances of hallucinations.
