Countries should own their AI task, decentralize deployment, open source is an important tool: Mistral AI CEO

“We have a lot of power in artificial intelligence. We really don’t want to be in a world where three or four enormous companies own the deployment and manufacture of AI… In a world where AI is going to generate multiple digits of GDP in the coming years, everyone running AI workloads needs to make sure they stop doing business. The global AI landscape doesn’t rely on external providers who can actually stop, but too often the narrative is shaped by a focus on control, surveillance and geopolitical leverage. India AI Impact He was speaking at the summit.

Noting that open source is “not a radical idea,” Mensch said it played a role in building cloud infrastructure and a secure Internet. He said there is a dichotomy in the AI ​​world between those who support open source and those who oppose it. “Some companies like Mistral are working to model the world’s knowledge, which can be used to create applications and provide services to organizations, states, education, health care and the rest of the world, where all these models are actually compressed by some large private corporations, which actually use them against their users,” said the CEO of an AI startup from France.

Given how AI is going to change the balance of the world in the next few years, Mensch says that “excessive leverage should be banned” to keep that balance stable and stable. “… and in that India is leading the way in betting on self-reliance, which is absolutely crucial for a country of this size,” he said.

Noting the diversity of India’s 22 languages, Mensch said there is a need to collaborate with local ecosystems and research institutes to create better open-source models and ensure that languages ​​feed into AI models. Mistral AI models are focused on making rare languages ​​more usable, he said. “We want to collaborate with companies and research labs that are working to get the content needed to improve the models,” he added.

The diversity of India’s culture and its ambition, the size of the market and the size of its industry give it enormous leverage when it comes to building AI, which is differentiated, becomes a major export product and gives them more leverage on the international scene, Mensch said. “By taking control of its own AI, India can become a global hub for innovation and lead the way for the rest of the world,” he added.

He said that governments can improve the quality of their services, the lives of citizens and the quality of life of communities. However, this will only work if everyone has access to AI and everyone understands how to use it. “That’s why multilingual AI is actually a very strong need. That’s why we need to focus on training people and bringing technology to them at an affordable price,” he said.

AI’s true potential lies in level-playing the field, Mensch said. To ensure this, he said, it is important that the technology being built does not widen the digital divide, ensuring that no language and culture is left behind. “We need to make sure the technology we’re building is affordable, and that’s why our focus on open source is a very important aspect,” he said, adding that governments should invest in their own AI infrastructure and ensure that it’s distributed to local and regional providers.