Tata Communications unfazed by US chip export curbs; in talks with enterprises for NVIDIA-powered data centre: MD

AS Lakshminarayanan, MD and CEO- Tata Communications

Despite the recent US government-imposed cap on AI compute and GPU exports, Tata Communications does not anticipate any immediate impact on its AI Cloud plans and is currently in a “wait-and-watch” mode to assess if the new Donald Trump administration introduces any potential changes.

“That was an interim rule that was passed. We've gone through the consultation. We will wait and see how it develops with a new administration coming in. Needless to say, I think the limits they talked about are quite high anyway, and that's over three years. So, I don't think there will be any immediate impact. Still, we'll have to wait and see how that order stands or if it gets modified with the new administration coming in,” MD and CEO AS Lakshminarayanan told Moneycontrol in an interaction.

Tata Communications is one of the major Indian companies aiming to offer AI cloud computing services to enterprises and startups using high-end NVIDIA Hopper and Blackwell GPUs. The company unveiled its AI infrastructure with NVIDIA-accelerated computing in October last year.

Moneycontrol on January 21 exclusively reported that India plans to raise concerns with the Trump administration about regulations that further restrict US exports of artificial intelligence (AI) chips and technology. The ministry of electronics and information technology (MeitY) has held discussions with various government departments and shared concerns over the new US rule and its impact on country's IndiaAI Mission.

The government is sourcing nearly 10,000 GPUs under the IndiaAI Mission, with the number potentially increasing in the future.

As per the new US directive, India won't be able to import more than 50,000 GPUs from the US in a year, and even at that level, India will require a license.

GPUs are critical for AI as they can handle multiple tasks simultaneously, significantly faster than regular computer processors. This speed is essential for training and running AI models, which require quickly processing large volumes of data.

Tata Communications’ GPU-as-a-Service platform went live last week and is powered by its data centre, which was set up in partnership with NVIDIA. The company has also been in discussions with several enterprise customers for onboarding.

“The entire GPU-as-a-Service offering was launched only a week ago. Until then, NVIDIA and we were engineering and setting up the data center and cloud infrastructure. So, we have just launched it, and it is on track. The full-fledged AI studio, which is the software platform on top of the GPU-as-a-Service, will help enterprises consume cloud capabilities more securely and efficiently. The AI studio platform will be launched later this year and is progressing quite well,” Lakshminarayanan added.

Tata Communications and Jio Platforms were among ten companies selected for the final bidding process to procure 10,000 GPUs under the IndiaAI Mission. The bidding for GPU procurement began on January 22, and results are expected within a week, Moneycontrol reported on January 22.

“MeitY had issued an RFP. The government intends to support startups with a scheme, and we are participating in that effort. We are very pleased to have reached this stage. Now, we need to see how it translates into business. We have to go through the subsequent stages, but we are pleased with the progress we have made,” A.S. Lakshminarayanan said regarding the IndiaAI Mission mega tender.

From India’s perspective, Lakshminarayanan emphasized that AI will play a critical role in business but cautioned that it is still in its early stages. Enterprises are primarily in an experimentation phase rather than large-scale deployment.

“In India, enterprises are also beginning to experiment and will move to the deployment stage later. This is a very evolving story, so we'll have to wait and watch rather than be overly concerned about any rules that are passed. We will remain watchful to see how enterprises adopt the capabilities that we are bringing to the market,” he concluded.