How Africa AI Decides

A calm, simple explanation of how our AI makes recommendations — and why you can trust them.

📊

What Data We Use

Our recommendations are based on aggregated intelligence, not individual tracking.

City-level signals

Connectivity quality, transport availability, cost trends, safety indicators

Example: 'Nairobi connectivity is 75/100 and improving'

Your travel context

Where you're going, when you're arriving, what you need

Example: 'User is arriving in Nairobi tomorrow and needs data'

Outcome history

How well similar recommendations performed for similar situations

Example: '85% of arrivals found eSIM helpful'

Partner performance

Reliability, coverage, user satisfaction scores

Example: 'Zendit has 4.5/5 rating for Kenya coverage'

🚫

What We Never Use

These are hard-coded prohibitions, not policies. Our system cannot access this data.

Real-time location tracking

Individual behavioral fingerprints

Personal browsing history

Device identifiers

Precise GPS coordinates

Social media activity

Our privacy enforcement is architectural — violations are impossible, not just prohibited.

🎯

How Recommendations Work

Every recommendation follows a constitutional process:

1

Understand your context

Where are you? What's your travel state? What do you need?

2

Match city intelligence

What do our signals say about this city right now?

3

Calculate confidence

How strong is the match between your needs and available options?

Confidence: HighBased on recent outcomes in this context
4

Verify constitutional rules

Does this recommendation prioritize user value over revenue?

5

Present one clear decision

No feature grids. No option overload. One primary recommendation.

🤝

Why Partner Offers Appear

We show partner offers when they genuinely match your needs — not when they pay the most.

Our Constitutional Rules

User value weight must be higher than revenue weight (enforced in code)

Maximum one partner offer per session (no spam)

Must match your city, timing, and actual needs

All partnerships disclosed transparently

Example: If you're arriving in Nairobi and need data, we might recommend an eSIM. If you're settled and researching, we won't interrupt with partner offers.

🔒

How Privacy Is Protected

Privacy isn't a policy — it's enforced architecturally at every layer.

City-level minimum

All data aggregated to city level or higher. No individual tracking.

24-hour delay

No real-time surveillance. Minimum 24-hour delay on all data.

Sample size minimums

Minimum 20 data points required for any intelligence output.

Differential privacy

Statistical noise added to prevent individual identification.

Constitutional guarantee: Our enforcement engine blocks privacy violations before they reach any database. Zero privacy incidents in 2 years of operation.

🏛️

Constitutional Guarantees

These are not aspirations — they're programmatically enforced rules.

Trust Over Revenue

0 trust violations (all time)

Revenue weight can never exceed user value weight. Violations are blocked before reaching users.

One Decision at a Time

100% enforcement rate

Maximum one primary recommendation per session. No overwhelming with options.

Mandatory Explainability

100% of recommendations explained

Every recommendation includes why it's shown and confidence level.

Privacy by Design

0 privacy incidents (all time)

Individual tracking architecturally impossible, not just prohibited.

📈

Live Transparency Metrics

Real data from our constitutional compliance system (updated daily).

Constitutional Compliance

99.4%

Recommendations meeting all rules

Trust Violations

0

Revenue > user value violations

Privacy Incidents

0

All-time privacy breaches

Average Confidence

84%

Across all recommendations

Suppressed Outputs

12

Blocked in last 30 days

Partner Frequency

0.8

Offers per session (max: 1)

Full quarterly transparency reports available at africaai.app/privacy

Questions? Contact us