The AI gold rush is unlikely to slow down

The AI gold rush is unlikely to slow down

AI has moved from technology labs into boardrooms, living rooms and even everyday conversations. As billions of dollars pour into building AI infrastructure, concerns are mounting over job losses, environmental costs and whether the investment frenzy is sustainable. The recent gains made by Democrats in the US midterm elections have fuelled hopes of greater regulatory scrutiny and a possible slowdown. But can political changes really put the brakes on a technological race driven by fierce competition between the USA and China? From the prospect of an AI bubble to India’s own ambitions, one thing seems clear: the AI gold rush is unlikely to slow down, Trump or no Trump.

Artificial Intelligence (AI) has become part of our everyday vocabulary. It is now a subject of boardroom deliberations, dinner-table conversations and even tea-stall chatter, given the profound impact it is having on our lives and its potential to disrupt the status quo.

With Democrats sweeping the US midterm elections this month, there are now voices hoping that the breakneck pace of AI proliferation will finally slow down. I, however, see this as wishful thinking. The beauty of technology lies in its inherently democratic nature: it transcends political boundaries, is not confined by geography and does not wait for political consensus.

While political control may appear to have shifted, the US can ill afford to slow down its AI ambitions when China is aggressively building its AI ecosystem and countries such as South Korea, the UK and India are strengthening their own capabilities. The fear of losing its technological edge to China or other emerging AI powers is likely to remain a powerful incentive for Washington to stay the course, irrespective of which party controls Congress.

Although automation has made people panicked in every era, the AI fear is palpable and we are already seeing companies’ laying-off employees in large numbers in the name of rightsizing and strengthening their books.

Not just at the governments or company level, ordinary people are also using AI in some form no matter how basic it is. Some are using it out of curiosity, others out of necessity or fear of losing relevance.

The AI picture
Billions of dollars are being invested in creating the AI infrastructure. According to a report published by Stanford Institute for Human-Centred Artificial Intelligence (HAI), AI capability is not plateauing but reaching far more people than ever.

The pace of investments in the US, especially by the hyperscalers like Nvidia, Alphabet (Google), Meta and Microsoft etc. is mind boggling. The overall investments in AI are expected to more than double in 2026 to USD 581 billion, according to estimates by Goldman Sachs. In 2025, the world’s largest economy witnessed investments to the tune of USD 285.9 billion.

Its research further estimates companies likely investing USD 1.019 trillion globally in AI this year.

The kind of money that has been going in this space has been a cause of worry for people across sectors. While the realization that automation is the only reality has been understood for long, with AI, the inevitability is coming much faster than it was estimated.

Trump upping the ante
It is a perception that Trump’s policy measures on AI were to create a US hegemony and to push its growth through policy measures and government support. AI is not just a software but requires massive infrastructure support.

The administration has done this through deregulation and removing restrictions on AI development. Trump on January 23, 2025, signed an executive order titled ‘Removing Barriers to American Leadership in Artificial Intelligence’. This was a U-turn from the policies of the previous administration.

The executive order sought to ease federal regulatory requirements for large AI data centres, accelerate environmental and other permitting processes. It further made provisions to sanction federally owned land available for data-centre development while facilitating construction of power infrastructure, including transmission lines, substations and power-generation facilities.

The approach showed a sense of urgency to take AI leadership, treating it as a strategic national-security and economic priority.

Why Democrats matter?
With the Democrats taking control of the Congress alongside several important governorships, it is being perceived that the government would find it hard to steamroll legislations. There is a fear that there will be more scrutiny of AI data-centre construction which could delay government clearances, hitting the investment cycle.

There are environmental concerns too as the same AI index report explains how AI’s environmental footprint has been increasing across power, water, and emissions, claiming Grok 4’s estimated training emissions reaching 72,816 tons of CO₂ equivalent in 2025.

Meanwhile, the AI data center power capacity rose to 29.6 GW, comparable to New York state at peak demand, and annual GPT-4o inference water use alone may exceed the drinking water needs of 1.2 million people.

The United States hosts 5,427 data centres, more than ten times any other country, consuming more energy than any other region, the report said.

While AI is a realty which cannot be put on the backburner, humanity needs to find a balance. The environmental concerns are real and cannot be wished away. AI’s proliferation is a global phenomenon and there is a price that every country will have to pay to promote it.

AI a bubble?
Even before the US midterm elections, the voices were erupting, dubbing AI boom a bubble. They questioned if the money chasing this sector is justified and if there will be quick returns on these investments (RoAs).

The magnificent seven viz. Google, Amazon, Apple, Meta Platforms, Microsoft, Nvidia and Tesla are the darling of investors, many of whom have invested out of the FOMO factor.

Investors are willing to pay substantial premiums for companies expected to dominate the next technological revolution while banking on the earnings visibility. Will that translate into actual gains is a hope, we are living with.

Some have even drawn parallels to the dotcom bubble of 1990s.

Off-late, there has been some anxiety among investors, many of whom have been offloading investments in these companies. Trump’s loss has further acted as a setback. The situation is not different in countries like China, South Korea and Japan.

Some are even likening it to the dotcom boom of the US in 1990s.

Why it is different this time?
AI’s penetration is far deeper and swift than what the dotcom period saw and is a global phenomenon now. It is riding on a strong internet ecosystem system which has been built over decades.

The internet startups then had no real revenues and much of the capex was funded through borrowing. In contrast, most of the AI infrastructure being built now is by technology companies with very deep pockets.

Dotcom came at a time when people were not used to technology and it had limited use case. The situation is starkly different now with smartphones in virtually every hand.

The companies are more innovative now to come out with one offering after another, that keeps users engaged.

While data-centre capacity, GPUs and power infrastructure are being built amid strong demand, future utilisation remains uncertain.

The AI boom is expected to stay.

Indian IT companies catching-up
While Indian IT companies have woken up late to the AI realty, they are now showing a sense of urgency to catch-up. They are fast scaling up investments in this area. Many companies have even created separate AI verticals and declaring revenues separately from this segment.

This is with an objective to change the narrative that they were caught napping when the world was burning mid-night oil.

There has been criticism of Indian IT companies lacking development acumen and happy being a service industry. The criticism is not unfounded as we do not have a single company that can match a Google, meta or an Apple.

Trump’s loss in the recently concluded polls have given a ray of hope to the Indian IT companies to catch-up when the global AI growth takes a breather.

However, it will be premature and even childish to sit back and relax as the likelihood of AI slowdown is a mere speculation.

In fact, we must use this break to build our own systems to cater to our domestic demands and also for the world instead of creating infrastructure to serve international companies.

We will need to integrate it at the policy level with big government support. At the same time, the private sector, especially the top 10-20 companies in India will have to look beyond servicing and focus on innovation and development. Time is running out!

Leave a Comment

Your email address will not be published.