sika-agent: open RL environment trains AI agents to catch MoMo scams in Twi and English
emeka_boris · x · 2026-10-17
Developer geeksilas released sika-agent, an open RL environment for training AI agents to refuse MoMo mobile-money scam calls and texts in English, Twi, and the Twi-English mix most people actually speak.
Results: Qwen3.5-9B completes 86% of tasks in English but only 51% in Twi, while frontier models stay at 90%+ across all languages (240 tasks per model). That gap is what training small local-language agents should close.
More from Safety
- Matthew Green highlights unexplained skipped curves in NIST P-256/P-224 seed generation — matthew_d_green · 2026-10-17
- AI tools reportedly helped one attacker hit 9 South Korean banks — Comfortable_Gene5180 · 2026-10-17
- Meta terms reportedly grant itself rights to scrape the open web for AI training — VeryWellVersed · 2026-10-17
- Jensen Huang on Lex Fridman: build the flywheel, don't just block chip exports — rohanpaul_ai · 2026-10-17
- METR: AI labs redacted key claims before approving frontier risk report — kevinnbass · 2026-10-17
- Macroscopic Ventures: Anthropic backer with stakes in Apollo Research, Gray Swan and Redwood — kevinnbass · 2026-10-17