When Micron Technology broke ground on its $2.75 billion facility in Sanand, Gujarat, the site had no water recycling infrastructure, no trained local workforce for cleanroom operations, and nothing resembling a supplier ecosystem within the state. The company had to negotiate water allocation in a region already under agricultural stress, coordinate utility connections with state agencies unused to semiconductor timelines, and start training engineers who had never worked inside a fabrication environment.
Nobody who has been through that process describes it as straightforward.
India Is Offering Real Money. The Small Print Matters.
Over $18 billion has been committed across 12 approved semiconductor projects as of early 2026. The India Semiconductor Mission offers capital subsidies, fast-track clearances, and direct matchmaking between global firms and Indian partners. Twenty four chip design projects have been approved alongside the fabrication commitments. India Semiconductor Mission 2.0, announced in the Union Budget 2026–27, extended those commitments further and a new investor portal now promises single-window support for land, utilities, and regulatory approvals.
The single-window system is an acknowledgment that the previous process was too fragmented for firms used to operating in more streamlined jurisdictions. Approval timelines have improved in states with strong political backing for semiconductor investment, particularly Gujarat. Whether that responsiveness holds as more projects come online simultaneously, and whether it extends to states with less experience handling foreign industrial investment at scale, are questions every incoming firm needs to answer through direct engagement rather than taking the policy documentation at face value.
What the portal promises and what a project team encounters six months into site preparation are often different conversations.
Gujarat, Uttar Pradesh, Odisha. Three States, Three Very Different Bets.
The investment map is broader than most coverage suggests and for foreign companies the geography carries real operational implications.
Gujarat has moved fastest. Existing industrial infrastructure, proximity to port logistics, and a state government with experience processing foreign industrial projects at scale made it the natural first choice for Micron. It remains the most developed location for semiconductor investment in India today.
Uttar Pradesh is the newer entrant. The HCL-Foxconn joint venture chose Jewar, outside Delhi, for its $446 million fabrication unit targeting 20,000 wafers per month. Proximity to the new international airport and the National Capital Region’s engineering talent pool drove that decision more than incentives alone.
Odisha is playing a longer game. Its $3 billion silicon-carbide plant announced in November 2025 and targeting production by 2028 is the most specialised bet in the current portfolio, built around power semiconductors for electric vehicles and renewable energy rather than general-purpose chip manufacturing.
The state selection question deserves more time than most companies give it. The incentive number is the wrong place to start. The more important questions are about bureaucratic culture, infrastructure maturity, and whether the state government has the institutional capacity to resolve problems after the groundbreaking ceremony is over. Gujarat has demonstrated that capacity. Other states are at earlier stages and the difference shows up in project timelines in ways that are hard to recover from once construction has started.
The Geopolitical Case Looks Clean Until You Look Closely
Foreign firms are not coming to India purely for subsidies. As US-China tensions have hardened, companies supplying American defence and technology markets face real pressure to move production away from China-linked supply chains. India offers democratic governance, a deepening technology partnership with Washington through the iCET framework, and a domestic semiconductor market the India Electronics and Semiconductor Association projects will grow from $52 billion in FY2025 to $103 billion by 2030. Those figures come from a body with an institutional interest in optimism and should be read as directional rather than precise. The growth trajectory is not seriously disputed.
SiFive and Western Digital have established chip design centres in India working on RISC-V processors for AI and embedded systems. Design activity typically precedes manufacturing investment which suggests the ecosystem is building from both ends simultaneously.
But there are complications in the geopolitical picture that belong in any serious analysis. India abstained on United Nations votes condemning Russia’s invasion of Ukraine. It maintains defence relationships with Moscow that long predate the current global realignment and shows no signs of unwinding them. Strategic autonomy is not a diplomatic formality in India. It reflects how the country actually navigates great power competition and it means India’s alignment with Washington has limits that will occasionally show up in ways foreign firms did not plan for. Building a supply chain strategy around the assumption of unconditional Indian cooperation with US trade policy is a bet worth examining carefully before making.
Eight Months. That Is How Long It Takes to Train One Engineer.
Back in Sanand, the training challenge Micron faced when it could not find cleanroom-ready engineers within the state is not a Gujarat problem. It runs through the entire industry.
India graduates around 600,000 electronics engineers a year. Only about one percent have the skills needed for fabrication and advanced packaging without significant retraining. A shortfall of between 250,000 and 300,000 trained professionals is projected by 2027.
Companies that have gone through semiconductor workforce buildouts in India describe retraining timelines of six to nine months for engineers coming from adjacent industrial backgrounds. Before those engineers are independently functional on a fabrication line. For someone hired without any prior cleanroom exposure the timeline is longer still.
That gap between projected and actual workforce readiness is where many India manufacturing projects have run into trouble. The timeline assumptions in most project models do not account for it and by the time the problem becomes visible the construction schedule has already moved on.
Tata Electronics noted at SemiCon India 2025 that even constructing a fab requires specialised skills India currently lacks at scale. The India Semiconductor Mission has partnered with 270 universities on chip design tools which logged over 12 million uses in 2025. A target of 500,000 trained professionals by 2030 has been set. The fabs being built now will need those people considerably earlier than that.
This Is an Assembly and Packaging Opportunity. Not a TSMC Moment.
India is not offering leading-edge fabrication. The realistic opportunity is assembly, testing, packaging, and mature-node manufacturing where global capacity is genuinely short and India’s cost structure is competitive. Every major chipmaking nation started here. South Korea and Taiwan did not begin at the leading edge either.
The entry point is coherent. What it requires is operational patience and a willingness to staff India seriously rather than treat it as a subsidiary decision. The firms that have struggled are largely those that underestimated the distance between signing an agreement and running a facility. That distance is longer in India than the incentive documentation suggests and shorter than the pessimists claim. It is navigable. But only if the people sent to navigate it are good enough to do so.
Micron’s experience in Sanand will be studied by every firm that follows. If it reaches meaningful operational capacity on something close to schedule it will do more for India’s semiconductor credibility than any policy announcement.
The Announcements End in 2028. The Real Test Starts There.
By 2028 the Odisha silicon-carbide plant should be operational and Micron’s Gujarat facility should be running at meaningful capacity. That is when the claims made in every announcement since 2022 will meet the reality of what India’s semiconductor ecosystem can actually sustain at commercial scale.
If Sanand is working and Odisha is on track, India will have a foundation that other states can replicate and other firms can build on. If the gaps between policy and delivery have proven wider than the incentives could bridge, that will be visible too and the investment calculus will shift accordingly.
For foreign companies evaluating India now the question is straightforward. Do you want to be part of the cohort that shapes what 2028 looks like, or the one that reads about it afterward.
