Google AI has introduced Science One, a novel framework designed to enable autonomous AI systems to conduct scientific research while maintaining transparency and verifiability through a chain-of-evidence methodology. The framework addresses a critical challenge in AI research: ensuring that autonomous systems can produce trustworthy scientific findings that can be independently validated and understood. By embedding evidence-tracking throughout the research process, Science One allows AI systems to document their reasoning, methodology, and supporting evidence at each stage. The chain-of-evidence approach represents a significant advancement in establishing accountability for AI-driven scientific discovery. Rather than treating autonomous research as a black box, the framework creates an auditable trail that enables researchers and stakeholders to examine how conclusions were reached and what evidence supported them. This is particularly important as AI systems become increasingly autonomous in conducting scientific experiments and analysis across fields ranging from materials science to biological research. The Science One framework reflects Google's broader investment in autonomous AI research capabilities and signals growing industry recognition that verifiability and transparency are essential prerequisites for deploying AI systems in high-stakes scientific domains. The work positions Google at the forefront of developing trustworthy AI research infrastructure, with potential applications across academia, industry R&D, and scientific institutions seeking to leverage AI while maintaining rigorous scientific standards.