Nvidia and US AI stocks crashed amid data-centre fears. See what it actually means for Infosys, TCS, Wipro and India’s data-centre linked names like L&T.
Wall Street had one of its worst days in months on 5 November, and this time it was not inflation or interest rates behind it. Nvidia fell over 9%, wiping out more than Rs 37 lakh crore ($440 billion) in market value in a single session, its second-largest one-day loss on record. The Nasdaq 100 lost roughly $2.3 trillion. Chip stocks as a group fell over 12%, and data-centre builders like Applied Digital and cooling specialist Modine were down double digits for a second straight day.
The trigger was not weak AI demand. It was politics. Democratic gains in the US midterms, including wins in states where opposition to large data-centre projects has been building over electricity costs and environmental concerns, raised fears that the AI infrastructure buildout could slow, get costlier, or face new regulatory hurdles. That is a very different kind of shock from the usual AI stock stories, and it matters where you look for the read-through to India.
It is worth being precise here, because two different AI-linked worries have been hitting markets this year, and conflating them leads to the wrong conclusions for Indian portfolios.
The first is the one from 5 November, a US-specific political risk to data-centre construction. This affects chipmakers, cooling and power equipment companies, and the "neocloud" builders that rent out AI compute capacity, almost all of which are American companies.
The second, and the one that has actually hurt Indian portfolios more directly through 2026, is the fear that AI tools themselves will disrupt the traditional IT services business model, fewer billable hours, compressed pricing, less need for large offshore teams. That worry has dogged Infosys, TCS, Wipro and peers through the year, independent of what happens to Nvidia on any given day. We covered this dynamic closely in why Indian IT falls even when global AI stocks rally, the contradiction explained, and it remains the more relevant story for anyone holding Indian IT stocks specifically.
Context matters here. The Nifty IT index has fallen sharply through 2026, following a double-digit decline in 2025 too. This has not been one single event, it has been a drumbeat of separate shocks. Accenture's guidance cut earlier in the year triggered a sharp one-day fall across Infosys, TCS and Wipro, something we broke down in why IT stocks slumped after Accenture's guidance cut and later revisited in whether Accenture's subsequent rally was a buy signal for Indian IT.
Separately, a wave of US work-permit restrictions hit sentiment on the sector too, which we covered in how IT stocks reacted after the PERM ban and the F&O setup that followed. And brokerages have grown more cautious along the way, a shift we tracked in CLSA turning cautious on TCS and Infosys and what changed. The common thread across almost all of these is the same underlying fear, that AI tools are changing how much work clients need Indian vendors to staff and bill for.
Indirectly, yes, but through sentiment and risk appetite rather than through company fundamentals. A sharp global tech selloff tends to spill into Indian IT ADRs overnight, and that spillover often shows up as weak opens on Dalal Street the next morning, even when nothing has changed about an Indian company's own order book. This is the connective tissue, global risk-off in tech drags down Indian IT shares even though Infosys and TCS have almost no direct exposure to US data-centre construction or chip spending.
There is also a second-order read. If the political pushback that triggered this selloff genuinely slows data-centre buildout in the US, that is arguably a mixed signal for Indian IT, slower AI infrastructure growth could mean slower AI-related disruption too, which is not entirely bad news for the traditional services model that Infosys, TCS and Wipro still depend on for the bulk of their revenue.
| Factor | US AI chip and data-centre stocks | Indian IT services stocks |
|---|---|---|
| Direct trigger on 5 Nov | Yes, political risk to US data-centre construction | No, almost no direct exposure |
| Main worry through 2026 | Is AI infrastructure spending sustainable | Will AI tools shrink demand for billable services |
| Channel of impact on India | Sentiment spillover via ADRs and global risk-off | Direct, company-specific revenue and pricing risk |
| 2026 performance so far | Volatile but still up strongly for the year before this selloff | Nifty IT down sharply, a weak year overall |
Unlike the US, India does not yet have large, pure-play, listed data-centre operators of the scale seen on Wall Street, but there is meaningful indirect exposure. Companies involved in building the physical infrastructure, land, power, cooling and engineering, around AI compute capacity in India stand to benefit from the broader buildout even if US-specific political noise rattles sentiment overseas. We covered one of the larger Indian AI-infrastructure deals in L&T's Rs 15,000 crore AI factory order with Nvidia and TogetherAI, and power-sector names tied to this theme in the PM Dhara scheme and the power, renewable and battery stocks it touches.
For a broader basket of names riding India's own AI and data-centre wave, our earlier roundup on five Indian stocks riding the AI data centre wave in 2026 remains a useful reference point, though it is worth watching whether global capex caution bleeds into domestic order timelines over the coming months.
Political-risk driven selloffs tend to be more reversible than structural demand worries. Nvidia's own guidance has consistently pointed to supply constraints rather than falling demand, and its next earnings report on 17 November will be the real test of whether this was overreaction or the start of something more durable. If the company's commentary reassures markets on actual order books, much of this specific selloff could unwind quickly, the way sharp AI-linked drops earlier in the year did, something we discussed when AI stocks lost tens of billions overnight on inflation fears in our piece on AI stocks losing $80 billion overnight.
The Indian IT disruption worry is a different animal. It is not a single headline event that reverses on one earnings call, it is a slow-moving question about how enterprise software budgets and staffing models evolve over several years. That is why Infosys, TCS and Wipro have moved more on quarterly guidance and large deal commentary through 2026 than on any single overnight US tech headline, even one as dramatic as this week's.
If your portfolio has direct exposure to Infosys, TCS, Wipro or HCLTech, the US data-centre selloff itself is largely noise for your fundamentals, watch Nvidia's 17 November earnings for sentiment, but do not treat it as a reason to change your Indian IT thesis. The more relevant inputs remain quarterly deal wins, large-deal commentary on AI-related work, and whether global capex caution from major clients shows up in guidance over the next couple of quarters.
If you are looking at India's own AI infrastructure theme, power, cooling, land and engineering names tied to domestic data-centre buildout, this US episode is a reminder that even that theme carries global capex-cycle risk, not just a straightforward growth story. Watching how US hyperscalers' capex plans evolve in the coming weeks will tell you more about the durability of that theme than any single day's price action in Mumbai or New York.
The selloff followed Democratic gains in the US midterm elections, which raised fears that new state and federal leaders could slow or restrict AI data-centre construction due to electricity and environmental concerns.
Not directly. Indian IT stocks have fallen through 2026 mainly due to fears that AI tools will reduce demand for traditional billable services, a separate and longer-running concern from the US data-centre political risk.
Not directly, since Indian IT majors have little exposure to US data-centre construction. The effect comes mainly through sentiment spillover via ADRs and broader global risk-off trading.
India does not yet have large pure-play listed data-centre operators at US scale, but companies in power, cooling, engineering and infrastructure have indirect exposure to the domestic AI buildout.
Nvidia's earnings on 17 November will be an important signal on whether actual chip demand remains strong despite the political uncertainty around US data-centre construction.