Research · Real World Impact

Measuring what AI changes for people.

Real World Impact research looks beyond the technology to the people and places it reaches: who benefits, who is left out, and how AI shapes work, health and opportunity across Africa.

Measured where people live.

Focus
People and communities
Projects
The AI Atlas · AI Economy 2035
Scope
Across Africa
Why it matters

Benchmarks say little about whether AI improves someone’s day.

We want to understand impact where it actually happens: in households, clinics, businesses and public services, and in the languages people use there.

This work combines listening with analysis. The AI Atlas gathers what people across Africa want from AI, and our economic research examines how AI could affect growth, employment and productivity.

Research areas

Where the work happens.

Real World Impact is made up of 3 research areas. Each has its own questions, methods and links to our products.

RWI-01

Reliable Agents in Real Conditions

Building AI agents that plan and carry out multi-step tasks reliably, and know when to ask a person before acting.

Real tasks are messy: information is missing, connections are slow and mistakes have costs. We study planning, memory and clear human-in-the-loop boundaries, so an agent can proceed when it is confident, ask when it is not, and stop when a task is outside what it should do.

  • Planning under uncertainty
  • Memory across tasks
  • Human-in-the-loop boundaries
RWI-02

Public Perspectives on AI

Listening to what people across Africa want from AI, and sharing what we hear openly.

Through the AI Atlas, people across African countries tell us where AI could help most and what worries them. We study how those views differ by country, sector and language, and use them to decide what to research and build.

  • The AI Atlas poll
  • Differences across countries
  • Open results
RWI-03

Economic & Social Impact

Examining how AI could affect growth, jobs and productivity across African economies, and who would gain or lose.

Our research on Africa’s AI economy explores scenarios to 2035. We look at which sectors and kinds of work are most exposed, where AI could raise productivity, and what skills, infrastructure and policy would help the benefits reach more people.

  • Scenarios to 2035
  • Work and skills
  • Sharing the benefits
Our approach

Measure what matters to people.

Impact research combines listening with evidence. We keep coming back to the people affected, before, during and after a system is in use.

  1. 01

    Listen

    Ask people and communities what they need and what concerns them.

  2. 02

    Define

    Agree on what a real improvement would look like, and how to measure it.

  3. 03

    Observe

    Study how AI is used in practice, in homes, clinics and businesses.

  4. 04

    Report

    Share what we find, including where things did not work.

The questions we are asking

What we are exploring.

01

What do people want from AI?

Through the AI Atlas, we ask people across African countries where AI could help most, and share the results openly.

02

How could AI shape economies?

Our research on Africa’s AI economy explores growth, employment and productivity scenarios to 2035.

03

Who benefits, and who is left out?

We examine how access, language and cost affect who can use AI, and where gaps appear.

04

How do we know a product helps?

Defining measures of usefulness, reliability and human agency for the products we build.

Where it connects

Related work.