HausaSafe

Nigeria

An open-source Hausa-language AI safety evaluation suite, closing a documented blind spot affecting 60M+ people currently invisible to every existing safety benchmark.

Project Details

I was translating an AI safety campaign poster, and words like "prompt" and "deepfakes" have no meaning in the Hausa Language. This is not a translation gap, but a safety infrastructure gap; AI can not be evaluated in a language whose safety vocabulary does not exist

I am building a reusable open-source set of AI safety test prompts in the Hausa language, which is the first of its kind that's documenting where frontier models fail, and why they fail, and what this failure reveals about assuptions embedded in AI safety discourse.

Theory of Impact

AI safety evaluations, red-teaming, benchmarks, and alignment testing are built almost exclusively in English. This isn't a minor gap; a model can pass every English-language safety benchmark while behaving unpredictably in Hausa, spoken by 60+ million people across northern Nigeria, Niger, and the Sahel, where AI systems are already deployed in health, finance, and security infrastructure with zero language-specific safety evaluation.

HausaSafe gives AI labs and safety researchers the tools to evaluate model behaviour in Hausa for the first time. This is grounded in the academic argument of Ireri, Abungu et al. (2026) that Africa is more likely to expose universal AI failure modes through distributional shift than to generate entirely distinct pathways of misalignment. This means Hausa evaluation doesn't just protect Hausa speakers; it surfaces failure modes the global safety community cannot currently find in English at all.

Ungoverned AI deployment at scale in low-resource, non-English-language contexts is an attack surface that the current safety architecture cannot see. Closing that blind spot, even partially, reduces the risk that catastrophic failure modes go undetected simply because nobody built the tools to look for them outside English.

People

Hadiya Usman

Team Member

Funding Details

Start Date: Aug 1, 2026
End Date: Oct 30, 2026
Expected Duration: 3 months
Funding Raised to Date: -
Annual Budget: -
Monthly Burn Rate: -
Current Runway: -
Funding Goal: -
Funding Stage: -
Fiscal Sponsor: -

Grants Received– no grants recorded

Add grant

Funding Asks

grantmaking.ai Launch Round

Applied

Minimum

$8,000

Ideal

$45,000

What this ask is

I was translating an AI safety campaign poster, and words like "prompt" and "deepfakes" have no meaning in the Huasa Language. This is not a translation gap, but a safety infrastructure gap; AI can not be evaluated in a language whose safety vocabulary does not exist

How the money will be spent

Minimum ($8,000) funds a focused 4-month sprint: fieldwork in Gombe State to validate the glossary and translation gaps with 10–15 community members and health workers using the completed KoboToolbox instrument, Hausa translation/review support, and publication of a citable dataset, the glossary, documented translation gaps, and first evaluation prompt set.

Ideal ($45,000) funds a full 12-month build-out: the above, plus a part-time technical developer to structure the evaluation prompts into a benchmark-compatible format AI labs can integrate directly, expanded fieldwork across multiple Gombe State communities, and initial work extending the same methodology to a second underrepresented language as proof that the model generalizes beyond Hausa.

Discussion

No comments yet. Be the first to share your thoughts.