TRUE, BUTSelf-Marked
Google is also the only one keeping score.
“Google DeepMind says Gemini Robotics ER 2 is its most capable embodied reasoning model, reporting 91.3% accuracy on timing benchmarks and the highest accuracy across core capabilities.”
Receipts pack
The evidence file for this verdict
3 receipts: 1 support the claim, 0 contradict it, 2 add context. 3 figures on the record, 2 traced to a quoted receipt.
Call resolution, recorded 28 Sept 2026
We called it. Checked on the due date, 60 days after the July 30 launch. A Google News search for Gemini Robotics ER 2 returned launch coverage and later features, none reporting an independent run of the 91.3% timing figure or the core-capability results, and an arXiv full-text search for Gemini Robotics-ER returned no paper submitted after the launch. On that record the call stands: no independent reproduction within 60 days. This is a search of the public record on 2026-09-28, not proof that no private test exists; the flips_if conditions still apply to any reproduction published later.
