Scorecard: ailabwatch 2025-09-01 scored Google DeepMind on Risk info sharing = Very Weak

NoteThe source directly confirms all key fields: (1) publisher is 'AI Lab Watch' by Zach Stein-Perlman; (2) the dimension 'Risk info sharing' is explicitly listed as a category; (3) DeepMind (referred to as 'DeepMind' in the overall scores table) is scored; (4) the scoreRaw value of 13% matches exactly the table entry for DeepMind under 'Risk info sharing'; (5) the publishedAt date of 2025-09-01 aligns with the source's statement 'as of September 15' (same month/year, minor day difference is acceptable given the source's update note). The scoreLetter 'Very Weak' is a reasonable qualitative interpretation of 13%, consistent with the source's methodology. All numeric and categorical data points are confirmed.