Asymmetric Cross-Lingual Transfer of Harm Representations (Igbo)

Testing whether the safety circuits in language models actually work in Igbo (Nigerian Language) and building evaluations that distinguish real refusals from accidental ones.

Endorsed by+1

Testing whether the safety circuits in language models actually work in Igbo (Nigerian Language) and building evaluations that distinguish real refusals from accidental ones.

Endorsed by+1

Project Details

I study whether the internal "harm direction" that mediates refusal in a language model transfers from English to Igbo, a language spoken by about 45M people in Nigeria. In Qwen2.5-1.5B-Instruct, using a 50-pair matched English-Igbo benchmark, I measured per-layer geometric overlap, cross-lingual AUROC transfer against same-language ceilings, and causal directional ablation.

Findings so far: the English-derived harm direction is at chance for Igbo at every layer. Ablating it at its point of peak overlap leaves Igbo refusal unchanged (12/12 before and after). The Igbo-derived direction transfers near-perfectly onto English. Many baseline Igbo refusals also look like generic comprehension failures rather than intent-aware refusals.

Status: submitted to Apart Research's Global South AI Safety Hackathon (June 2026) and selected for Apart Lab Studio. Limitations: one small model, 50 pairs, unaudited translations, and ablation at a single layer. A possible topic-level confound is still untested.

Next: run the complementary ablation, audit translations with native speakers, add models and larger datasets, extend to Yoruba and Hausa, and publish an arXiv preprint.

Theory of Impact

Safety training and evaluation are mostly done in English, yet deployed models serve hundreds of millions of people in lower-resource languages. If safeguards are not learned in those languages, refusal rates there can look reassuring while reflecting comprehension failure, not protection. That means (1) safeguards may be bypassed by switching language, and (2) evaluators may certify models as safe on misleading evidence.

This project builds methods that separate real, intent-aware safety behaviour from accidental refusals, and produces evidence on whether safety mechanisms generalise across languages. The intended chain is: better multilingual safety evaluations, then developers and evaluators catch language-switch failures before deployment, then fewer models ship with safeguards that only work in English. This is an indirect route to catastrophic-risk reduction through more reliable misuse safeguards and evaluation, and I'm not claiming more than that.

People

Chijioke Ubajaka

Team Member

Grants Received– no grants recorded

Add grant

Funding Asks

Funding ask

Open

Minimum

$6,000

Ideal

$15,000

How the money will be spent

Ideal ($15,000)

Compute: $4,000. GPU time to replicate on larger open models (7-8B class and above) and to run the complementary ablations.
Translation audit and dataset expansion: $3,500. Native-speaker review of the Igbo translations, then a larger matched benchmark, extended to Yoruba and Hausa.
Researcher time: $7,500. Part-time work to run the experiments, analyse the comprehension-failure refusals, and write the preprint.

Minimum ($6,000)

Compute: $1,500. One additional model and the complementary ablation.
Translation audit: $2,000. Native-speaker review and expansion of the Igbo set only.
Researcher time: $2,500. Analysis and preprint write-up.

Discussion

RRobert.amanfuEndorsement2d

A great project that tackles an underserved area of “low” resource languages. Looking forward to see where this research goes!

Chijioke Ubajaka2d

Hello @Robert.amanfu thank you for your endorsement!