How we work

Five values for how we work

Use these values to see what Data-Driven AI expects from people, leaders, and AI-assisted work.

The whole set

A value only matters when it changes a choice

Our five values cover the full path from taking a task to checking the result. They also set a clear limit: AI may help with the work, but a person still owns the outcome.

Values in practice

What each value asks us to do

Take ownership

Name the owner, carry the work through, and close the loop.

Keep learning

Ask questions, test ideas, and use mistakes to improve the next attempt.

Become an agent leader

Focus people on useful outcomes. Give suitable tasks to agents, then check the work.

Act with an AI-first mindset

Test whether AI can help. A person must refine the result and own the decision.

Be kind

Consider people, the planet, and the effect of the work on both.

Human control

AI does not remove responsibility

The person using an AI output remains responsible for checking it and deciding what happens next.

ChooseUse AI only where the task and risk allow it.

CheckReview facts, sources, access, and likely effects.

OwnKeep a named person accountable for the result.

Kindness in practice

Care needs evidence as well as intent

Our climate page separates a past reporting period from current claims. It shows the boundary, footprint, offsets, and targets in the published record.

Read the climate record