Index
Scorecard: saferai 2025-10-01 scored Google DeepMind on Risk Governance = Very Weak
Verdictconfirmed95%
1 check · 5/18/20261 → confirmed
Our claim
entire record- Snapshot
- saferai-2025-10
- Entity
- Google DeepMind
- Dimension Slug
- risk-governance
- Dimension Label
- Risk Governance
- Dimension Weight
- 0.25
- Score Numeric
- 12
- Score Letter
- Very Weak
- Score Raw
- 12%
Source evidence
1 src · 1 checkconfirmed95%Haiku 4.5 · 5/18/2026
NoteThe source is a SafeRAI scorecard table comparing frontier AI companies on risk management criteria. The table clearly shows Google DeepMind in the Risk Governance (section 4) row with a score of 12%. This matches all key fields in the record: publisher (saferai), publishedAt (2025-10-01, confirmed by "Up to date as of October 2025"), entity (Google DeepMind), dimension (Risk Governance - section 4), and scoreRaw/scoreNumeric (12%). The letter grade 'Very Weak' is a reasonable interpretation of a 12% score on a percentage scale. All verifiable data points align with the source.
Case № saferai-2025-10|sid_A4XoubikkQ|risk-governanceFiled 5/18/2026Confidence 95%