International Business Machines (IBM) announced on Tuesday its plan to release a suite of artificial intelligence models as open-source software. Additionally, the tech giant will assist Saudi Arabia in training an AI system in Arabic, marking a significant step in its AI strategy.
Unlike competitors such as Microsoft, which closely guard their AI models and charge for access, IBM is opting for a more open approach. Following in the footsteps of Meta Platforms, IBM is providing open access to its AI models. On Wednesday, the company introduced its “Granite” family of AI models, allowing businesses to customize these tools to suit their needs.
The Granite models are designed to expedite the process of writing computer code for software developers. IBM plans to monetize these models through a paid tool called Watsonx, which ensures smooth operation of the models within data centers post-customization.
IBM’s revenue strategy focuses on generating income when customers effectively utilize the AI models, regardless of whether the models originated from IBM or other sources, and whether they run in IBM’s data centers or elsewhere.
“We believe we’re in the early days of generative AI models,” said IBM Chief Executive Arvind Krishna in an interview with Reuters. “Competition at the end of the day is good for buyers. We also want to be safe and responsible.”
Furthermore, IBM announced a collaboration with the Saudi Data and Artificial Intelligence Authority to train its “ALLaM” Arabic language model using Watsonx. This partnership aims to enhance IBM’s AI capabilities by adding comprehensive Arabic language skills, including the ability to understand various Arabic dialects.
