What changed
Google expanded limited early access to its Geospatial Reasoning agent through Ask Google Earth, a headless Agent-to-Agent service and a Planetary Prediction Engine. It also released a technical report and development splits for two Google-created geospatial benchmarks.
Why it matters
The update turns a research prototype into a testable specialist-agent stack for multi-step spatial analysis and model training. It also supplies benchmark artifacts, while keeping the actual service experimental and access-controlled.
What to do
Apply for early access only with an expert-owned workflow. Reproduce one benchmark task, inspect the plan and source data, compare the agent with a deterministic GIS baseline, and require a domain expert to approve outputs before operational use.
Caveat
Access is limited and experimental. Google authored the system, report and two new benchmarks; one headline benchmark uses an autorater against workflow steps rather than final-answer correctness, so vendor scores are not independent proof of field reliability.