Primary document

'ML Fairness Communication Best Practices' (privileged)

Internal guidance on how employees should write about fairness and bias in Google's algorithms, including when and how to keep such documents under attorney-client privilege.

Internal Google material released by former engineer Zach Vorhies in 2019

Employees' email addresses, phone numbers and most names have been blacked out. Executives and document authors already named in public coverage are left visible.

Download the redacted PDF

'ML Fairness Communication Best Practices' (privileged), page 1

'ML Fairness Communication Best Practices' (privileged), page 2

'ML Fairness Communication Best Practices' (privileged), page 3

'ML Fairness Communication Best Practices' (privileged), page 4

'ML Fairness Communication Best Practices' (privileged), page 5

'ML Fairness Communication Best Practices' (privileged), page 6