Google DeepMind — Overdue

0 Overdue now

0 Upcoming

0 Missed

1 Met

0 Partial

1 Pending

100% kept (1/1 resolved)

kept = met ÷ resolved (resolved = met + missed + partial; counts shown so the number always carries its context).

Google DeepMind evaluations ⚠ under review Pending

Evaluate models at a compute / fine-tuning cadence

DeepMind’s FSF v1.0 stated an aim to evaluate models for every 6x increase in effective compute and every three months of fine-tuning progress.

Awaiting every 6x effective compute / 3 months fine-tuning (FSF v1.0)

Sources: Google DeepMind ↗ archived · auto-checked daily · reviewed 2026-06-18

Obligation (summarized)

“every 6x increase in effective compute and every three months of fine-tuning”

Why this ruling

This 6x / 3-month wording is v1.0 language; FSF v2.0 (2025) replaced the specific numbers with more flexible criteria.

Google DeepMind safety-framework ⚠ contested ⚠ under review Met

Implement the Frontier Safety Framework by early 2025

DeepMind’s Frontier Safety Framework v1.0 (May 2024) aimed to have the framework implemented by early 2025; FSF v2.0 was published on 2025-02-04.

resolved 25 days early

Sources: Google DeepMind (FSF v1.0) ↗ archived · Google DeepMind (context) ↗ archived · auto-checked daily · reviewed 2026-06-18

Committed text

“implemented by early 2025”

Why this ruling

FSF v1.0 stated an aim to have the framework "fully implemented by early 2025"; v2.0 (2025-02-04) specified the promised protocols and capability levels. Whether publishing v2.0 fulfills a commitment to "implement" is debatable; "early 2025" encoded as a 2025-03-01 checkpoint.