Deep Dives collects longer analysis of AI products, model behavior, technical changes, and strategic shifts. The hub is designed for readers who want more context than a news brief and need durable explanations for AI decisions.
20 articlesDeep Dives collects longer analysis of AI products, model behavior, technical changes, and strategic shifts. The hub is designed for readers who want more context than a news brief and need durable explanations for AI decisions.
Use this hub to follow the latest related articles, then open individual posts for source-grounded details, context, and practical implications.
Google Research just dropped two companion studies in Nature Cancer on using AI in NHS breast cancer screening. Large-scale retrospective evaluation plus a prospective…
Google researchers tested six LLMs on expert-level high-temperature superconductivity questions. NotebookLM and a custom system scored best, but all had key weaknesses.
Google's ConvApparel dataset exposes how LLM-based user simulators fail to mimic real humans—they're too patient, too knowledgeable, and too agreeable. Here's why that matters…
Google Research introduces a framework that adapts psychological questionnaires into situational judgment tests to measure how well LLMs align with human behavioral dispositions. Testing…
Google Research challenges the standard 1-5 rater approach in AI benchmarks, showing that depth over breadth often yields more reproducible results. Their new framework…
Google's ReasoningBank framework lets agents distill generalizable reasoning strategies from both successes and failures, moving beyond simple action logs to higher-level tactical insights.
Google Research's Simula framework treats synthetic data generation as mechanism design, using reasoning to build datasets from first principles. It gives fine-grained control over…
Google Research's new MoGen model generates synthetic neurons to train AI, cutting reconstruction errors by 4.4% and saving hundreds of person-years in manual proofreading…