HerWILL datathon draws participants from 45 countries
HerWILL has wrapped up the AI for Digital Safety Datathon 2026, a global initiative tackling harmful online content through responsible, context-aware artificial intelligence.
The competition attracted over 2,167 participants from 45 countries and 60 institutions worldwide, HerWILL said.
Participants represented countries including Egypt, Bangladesh, Syria, and Senegal.
Azra Tuni of Bangladesh was crowned champion, with Nour Mohammad of Syria and Syed Zayan Anwar of Bangladesh taking first and second runner-up positions, respectively.
They were chosen from seven finalists representing Bangladesh, Syria, Egypt, and Cameroon.
The datathon challenged participants to develop solutions for detecting harmful online content—explicit, subtle, or neutral—particularly in multilingual and low-resource contexts.
The programme began with a seven-day workshop series covering machine learning, natural language processing, and applied deep learning.
Participants then applied their learning in a competitive environment, using a real-world, human-annotated dataset.
The initiative was designed and implemented by STEM Lead Saiyara Mahmud and Co-Lead Tanmoy Mazumder, under the supervision of HerWILL Founder and CEO Farhana Hasan.
The programme emphasised recognising subtle toxicity beyond binary classification, addressing code-mixed language challenges in English, Bangla, and Banglish, and evaluating models using the F1NOP metric to balance performance and efficiency.
It also contributed to building a large-scale multilingual dataset of nearly 60,000 annotated samples to support upcoming research in AI safety.
The initiative was led by HerWILL in collaboration with global partners.
Strategic partners included the National Science Foundation and the AI Institute for Advances in Optimisation. Premium partners were SmartBizPay, Build Bangladesh, Galaxy for Development, and AMLD.
Further support came from the UIU Data Science Club, IUT Career and Business Society, and Government Titumir College Data Science Club, with professional development from BYLC.
The Business Standard served as the print and media partner.
HerWILL said the initiative demonstrates how inclusive, globally connected learning models may generate meaningful AI solutions and equip a new generation to design more protected digital systems.
