Prime Minister Narendra Modi held discussions with Mistral AI chief executive officer Arthur Mensch on the evolving global artificial intelligence (AI) landscape, focusing on trusted systems, innovation and the need to keep AI human-centric and inclusive.

“Talked about prospects for partnerships in India in diverse aspects relating to AI. India remains committed to developing AI solutions that empower humanity while promoting innovation, trust and international cooperation,” Modi wrote on X.


Mistral AI is a France-based AI company founded in 2023 that develops large language models and AI systems, focusing on open-weight architectures and enterprise deployment.

The startup is in talks to raise around €3 billion ($3.5 billion) at a valuation of roughly €20 billion, per a report by Reuters on Wednesday.

The remarks came on the final day of the three-day G7 summit hosted by France, which gathered leaders from Canada, France, Germany, Italy, Japan, Britain and the United States, along with invited guests including Brazil, India and Kenya.

Tech leaders, including those from Anthropic, OpenAI and Google attended a working lunch on Wednesday at the G7 to discuss regulation and AI infrastructure.

Mensch told ANI that his discussion with Modi focused on making AI socially accessible and advancing “sovereign AI capabilities,” including building full-stack systems that remain under national control.

Speaking about the meeting, he described it as a “great pleasure” and said the talks covered digital autonomy, the importance of enabling countries to “own and control the stack of artificial intelligence,” and expanding access to AI for citizens, civil servants and businesses.

Mensch expressed strong interest in collaborating with Indian companies to expand AI capabilities in areas such as startup ecosystems and large-scale public training to improve productivity and adoption.

Model launch underway

Mensch, in separate public remarks, outlined the French AI company’s development roadmap, including plans to release a new model later this summer.


He described it as “the start of a new family of models,” adding that it would be “fat indeed, but sparse,” indicating a shift in architecture and capability.

He said an early access programme for the model will begin in July for selected partners across research, government and industry.

Also Read: Mistral sees AI as utility, emphasis more on efficiency: Founder Arthur Mensch

Open-weight models are the way forward

According to Mensch, the upcoming systems will be “open-weight,” a design choice he said is important for transparency and user trust. “We believe this is critical for our customer confidence and for the research and developer communities,” he said.

The launch aligns with the company’s commitment to top one billion euros ($1.2 billion) in revenue this year, as CEO Mensch said in January at the World Economic Forum in Davos.

Mensch also detailed the company’s infrastructure strategy, stating that its deployment and training platforms are portable and can operate within a user’s virtual private cloud, on-premise data centres or infrastructure managed independently of US service providers.

He added that the company is expanding computing capacity and working with governments and enterprises globally to support AI systems that operate with greater autonomy and control.


'AI is the new oil'

Mensch characterised AI as a strategic global resource, stating: “AI, just like oil in the 20th century, is about to become the major source of leverage and power in the world.” He added that the trajectory of AI development could lead either to “a world of wealth and abundance for all” or to “the worst extractive economies that the world has ever seen,” depending on how the technology is governed and deployed.

Also Read:India can become a global AI innovation hub: Mistral CEO Arthur Mensch