"The AI industry in Egypt... doesn't exist. So he built his own: Horus, named after the ancient Egyptian god of the sky."
"Models are fluent in English and, to a lesser extent, Chinese, but far less capable in most other languages. So-called minority languages are, in reality, spoken by the global majority."
"Training AI models is expensive, and companies have little incentive to build for smaller language groups without a clear return."
"The rise of local LLMs, along with big AI companies tightening token limits, has opened space for smaller players."
Assem Sabry developed Horus, an AI model representing Egyptian culture, due to the absence of such models in Egypt. He aimed to reduce reliance on American and Chinese models by creating a more localized alternative. Horus was trained using GPUs from Google Colab and open-source datasets, achieving over 800 downloads in its first week. The AI industry has historically favored English and Chinese, leaving many languages underrepresented. However, the emergence of local LLMs is beginning to change this dynamic, allowing for more diverse AI development.
Read at Fast Company
Unable to calculate read time
Collection
[
|
...
]