Kenya → Africa ai@b4life.institute
B4Life.Institute
B4Life.Institute Discovery | Learning | Innovation
This week's BSI Weekly field
Live

Who governs the algorithms that govern us?

Digital public services, policy, and AI accountability.

8 research themes Certificate programme open Expert contributions welcome
Why this field matters

The state is being written in code.

Governments are being transformed from reactive administrators into data-informed, algorithmic systems. Digital identity, AI-driven service delivery, predictive policy, and open data platforms are reshaping the relationship between states and citizens — often before the institutions meant to govern them have caught up. The question this field asks is not whether this transformation is happening, but whether it will expand inclusion and public trust, or entrench the inequalities and exclusions that already shape who is served and who is overlooked.

Start here

The three questions

For any digital governance system, ask: Who is it accountable to? Who can contest its decisions? Who controls the infrastructure it runs on? If any of the three answers is unclear, the system is not yet governable.

What the field argues about

The five dilemmas.

These are the tensions practitioners in this field face — and the questions that shape its research agenda and its programmes.

01

Who is invisible to the algorithm?

Digital identity, algorithmic service delivery, and predictive policy all assume a legible citizen — one who is registered, documented, and consistently addressable. The people most in need of public services are often the least legible to the systems designed to serve them.

02

Accountability without comprehension.

When a decision about a benefit, a permit, or a policing priority is made by a system nobody can explain, the citizen has no meaningful way to contest it. Algorithmic governance without algorithmic accountability is administration without appeal.

03

Sovereignty over the stack.

States are increasingly running on infrastructure, models, and platforms they do not control. Data sovereignty, secure identity, and independent evaluation are the difference between a government that governs and one that administers someone else's system.

04

Trust as infrastructure.

Public trust is not a byproduct of digital governance; it is a precondition. Systems that succeed without it are technical exercises. Systems that fail with it are political failures. The field's hardest problems are questions of legitimacy, not engineering.

05

The capacity gap.

Digital transformation requires capabilities — procurement, evaluation, legal, technical, and civic — that most governments do not yet have. Without them, the field becomes a catalogue of purchased platforms rather than a shift in how states work.

Where to go next

Enter the field.

This field is built in four parts. Each one is a door.