‘I think you’re testing me’: Anthropic’s new AI model asks testers to come clean

Anthropic has unveiled its latest AI model, Claude Sonnet 4.5, which showcases advanced capabilities in understanding user intentions during testing. This development is significant as it raises important questions about the safety and reliability of AI interactions, suggesting that previous models may not have been as transparent. The release of a detailed safety analysis highlights the company's commitment to responsible AI development, making it a noteworthy advancement in the field.