1. Anthropic, an AI research company, is seeking to fund the development of a new and more comprehensive set of AI benchmarks. These benchmarks are intended to provide a more robust and thorough evaluation of AI systems, going beyond the limitations of existing benchmarks.

2. The goal of these new benchmarks is to assess a wider range of AI capabilities, including reasoning, common sense understanding, and the ability to handle complex, real-world tasks. Anthropic believes that current benchmarks often fail to capture the full breadth of AI capabilities and that a more comprehensive set of tests is needed.

3. Anthropic is inviting researchers and AI developers to contribute to the development of these new benchmarks. The company is offering funding and resources to support this effort, with the aim of creating a more rigorous and reliable way to evaluate the performance and capabilities of AI systems across a variety of domains.