Orange announced on Tuesday it is expanding its collaboration with OpenAI to deploy two new advanced open-weight reasoning models in its data centre infrastructure to boost support for AI solutions development across its footprint, particularly in Africa.
Orange said the OpenAI models, ‘gpt-oss-120b’ and ‘gpt-oss-20b’, enables the operator to flexibly deploy them across a variety of environments in its data centres with greater control. That will allow Orange to host AI workloads locally across any of its 26-country footprint, whilst safeguarding sensitive data and complying with diverse and evolving national regulations across Europe, Middle East, and Africa.
One core focus of the project is to accelerate Orange’s current collaboration with OpenAI to incorporate regional African languages into large language models (LLMs). Orange said that by accessing OpenAI’s gpt-oss model and fine-tuning them with local languages, customers across its 18-country footprint in Africa wil be able to communicate naturally in their local languages with customer support, sales and marketing.
Orange also said it intends to release the customized AI models in open source for free to local government authorities to use across their public services.
Meanwhile, the deployment of OpenAI’s open-weight reasoning models in Orange’s data centres will also contribute to development of numerous innovative applications and services for Orange customers, from natural-language voice assistants and chatbots to smarter network automation and resilience, as well as business solutions such as such as AI chatbots, advanced voice recognition-based services, and AI analysis of sensitive data tailored to specific customer needs.
“This strategy drives new use cases to address sensitive enterprise needs, help manage our networks, enable innovating customer care solutions including African regional languages, and much more,” said Orange’s chief AI officer Steve Jarrett in a statement.
