Construction is underway, billions of dollars are committed, and Tesla has ordered most of the initial equipment for its experimental semiconductor factory. But can Elon Musk’s companies accomplish in two years what established chipmakers spend decades perfecting?

Elon Musk is attempting something that few companies in the world have the financial resources, technical expertise, or ambition to pursue.

Through a joint initiative between Tesla and SpaceX, Musk plans to build one of the world’s largest and most advanced semiconductor manufacturing operations.

The project is called Terafab.

Its objective is extraordinary: manufacture enough advanced artificial intelligence chips every year to support approximately one terawatt of computing power.

Those chips are intended to power Tesla’s autonomous vehicles, humanoid robots, artificial intelligence systems, and potentially even SpaceX’s proposed orbital data centers.

But the scale of the project is only part of the story.

In October 2026, Tesla and SpaceX disclosed important new details about Terafab’s construction progress, equipment procurement, manufacturing schedule, and anticipated costs.

Tesla has already started building an experimental chip fabrication facility in Austin, Texas. SpaceX has begun developing the site of a much larger production complex in Grimes County.

The first phase represents more than $16.8 billion in announced investment.

County records reviewed by Engineering News-Record outline a potential multi-phase expansion reaching an astonishing $119 billion.

And during SpaceX’s October 20 earnings call, Musk revealed that the companies hope to manufacture their first functional test chip by May or June 2027.

Full-scale chip production is targeted for late 2028.

The central question is no longer whether Terafab is a real project. It is whether Tesla and SpaceX can master one of the world’s most technically demanding industries fast enough to justify the investment.

What Is Tesla and SpaceX’s Terafab?

Terafab is a proposed vertically integrated semiconductor manufacturing complex designed to bring several critical stages of chip production together.

Traditional semiconductor manufacturing involves a network of highly specialized businesses.

A company might design processors in California, manufacture them through a foundry in Taiwan, source memory from another supplier, and use additional facilities for packaging and testing.

Terafab aims to reduce those dependencies.

Its planned operations encompass chip design, photomask production, wafer fabrication, advanced packaging, and system-level integration.

In theory, that could allow Tesla and SpaceX to develop specialized chips, manufacture them, test the results, improve their designs, and repeat the process more quickly.

The companies believe existing chip suppliers will struggle to expand production sufficiently to support their ambitions for AI, robotics, and space computing.

Tesla needs processors for its Full Self-Driving systems and Optimus humanoid robots.

SpaceX envisions specialized processors powering enormous AI computing systems, including future facilities in orbit.

Their official Terafab website identifies three intended product families:

  • AI5: Advanced processors for Tesla’s driving systems and Optimus robots.
  • AI6: A subsequent generation of processors intended for increasingly capable robotics and AI workloads.
  • D3: Specialized processors designed for space-oriented computing applications.

Whether all these products ultimately achieve commercial-scale production remains to be seen.

But the strategy is clear.

Musk wants his companies to control more of the technology required to manufacture intelligence at enormous scale.

Two Semiconductor Factories, Two Different Missions

One of the most important details in Terafab’s development is that it involves two distinct manufacturing sites.

Understanding the difference is essential to evaluating the project’s progress.

1. Austin: The Experimental Research Fab

In April 2026, Tesla broke ground on a semiconductor research facility at the north campus of Gigafactory Texas.

This facility is intended to establish and test manufacturing processes before those techniques are transferred to a much larger production operation.

On Tesla’s October 21 earnings call, CFO Vaibhav Taneja confirmed that the company had already placed most of the initial equipment orders for the research fab.

Musk said machinery commissioning is planned for the first half of 2027.

The company hopes to produce a functional test chip around May or June of that year.

Initially, that chip may contain relatively simple static random-access memory, or SRAM, circuitry.

SRAM is a form of memory commonly used in processors.

Creating a test chip allows engineers to determine whether essential fabrication steps are working.

However, producing one functioning test chip is very different from manufacturing advanced processors at high volume.

The research facility must demonstrate repeatable processes, manageable defect rates, and consistent results across many wafers.

That is where the real technical challenge begins.

2. Grimes County: The Massive Production Facility

On August 6, 2026, Texas Governor Greg Abbott announced that SpaceX would construct the primary Terafab manufacturing complex in Grimes County, east of College Station.

The first phase represents more than $16.8 billion in proposed capital investment and is expected to create approximately 3,000 jobs.

The ultimate development concept includes roughly 100 million square feet of manufacturing facilities.

For perspective, that is an extraordinary industrial undertaking—even compared with major automotive factories and semiconductor campuses.

Site preparation is already underway.

An October 6 aerial survey documented extensive grading, construction access roads, and early groundwork.

But there is an important limitation to what these photographs prove.

The Grimes County site is being prepared for construction. It has not yet demonstrated the ability to manufacture advanced semiconductors.

Building cleanrooms, installing equipment, commissioning machinery, and producing commercially viable chips are separate milestones.

Elon Musk’s Aggressive Timeline: First Chips in 2027, Production in 2028

During the October 20 SpaceX earnings call, Musk provided the clearest public schedule yet.

His targets are:

MilestoneTarget
Initial equipment commissioningFirst half of 2027
First functional test chipMay–June 2027
Grimes County volume productionThird or fourth quarter of 2028
Long-term terawatt-scale outputNo established completion date

Musk acknowledged that the timeline is highly aggressive and that the companies might not achieve it.

That warning matters.

Advanced semiconductor fabrication requires extraordinary precision.

A modern processor may contain billions of transistors built through numerous repeated stages of deposition, lithography, etching, cleaning, and inspection.

Each manufacturing step must function within incredibly narrow tolerances.

Small imperfections can render chips unusable.

Even established chipmakers routinely spend years refining new manufacturing processes and improving production yields.

Tesla and SpaceX are attempting to establish those capabilities while simultaneously developing an enormous new industrial campus.

The 2028 target should therefore be treated as an ambitious internal objective, not an assured production date.

Intel Is Involved. TSMC Is Interested. But Major Questions Remain.

Terafab may be Musk’s vision, but it cannot be built entirely using expertise developed inside Tesla and SpaceX.

The companies need advanced semiconductor technology, specialized manufacturing equipment, and experienced engineers.

That makes their relationships with established chipmakers particularly important.

In April 2026, Intel announced its involvement in the project.

Intel brings decades of semiconductor manufacturing experience, including process engineering, fabrication technology, and advanced packaging.

Its participation potentially reduces some of the technical obstacles facing Terafab.

However, the precise commercial arrangement and division of responsibilities have not been fully disclosed.

Taiwan Semiconductor Manufacturing Company, better known as TSMC, is another potential participant.

On October 15, TSMC CEO C.C. Wei confirmed that the company had evaluated possible cooperation in Texas and discussed the Terafab opportunity.

But negotiations remain unresolved.

Wei indicated that facilities using TSMC’s manufacturing expertise would need to be operated by TSMC.

Musk has expressed a different preference: Tesla and SpaceX intend to build and operate Terafab themselves.

This distinction is far more than a disagreement over management titles.

Operating an advanced semiconductor fab requires proprietary manufacturing knowledge developed over decades.

A partnership in which TSMC operates manufacturing lines would involve a fundamentally different arrangement from one in which Tesla and SpaceX independently control the factory.

The final structure could materially influence the project’s risk, ownership, economics, and speed of development.

ASML and the Semiconductor Equipment Bottleneck

Perhaps the most important supplier to watch is ASML.

The Netherlands-based company manufactures the extreme ultraviolet lithography systems used in leading-edge semiconductor production.

These systems are among the most sophisticated industrial machines ever developed.

They enable chipmakers to create extremely small and complex circuit patterns.

Without the appropriate lithography equipment and complementary manufacturing systems, a company cannot simply begin producing competitive advanced processors.

During the October 20 earnings call, Musk thanked ASML and other suppliers for accelerating equipment deliveries.

But when an analyst asked whether SpaceX had placed orders for EUV systems, Musk did not directly confirm the order status.

That leaves a crucial distinction.

Tesla has placed most of the initial equipment orders for the smaller research facility.

The precise procurement and delivery status of every critical machine required for the large production facility is not yet publicly established.

For investors following Terafab, equipment installation and successful process qualification will be more informative than broad announcements of partnerships.

Could Terafab Ultimately Cost $119 Billion?

The project’s financial scale may be even more extraordinary than its technology.

Texas announced more than $16.8 billion for the first development phase.

But in October, Engineering News-Record examined public records indicating that the project could eventually involve as much as $119 billion across four construction phases extending through 2036.

The records also contemplate approximately $55 billion in investment for initial phases.

These figures are not interchangeable.

The $16.8 billion figure concerns the announced first phase.

The higher amounts reflect possible expansions and do not establish that every future stage has been authorized, financed, or committed.

The records reveal that Terafab’s infrastructure requirements extend well beyond semiconductor manufacturing equipment.

Plans include:

  • Specialized semiconductor cleanrooms and fabrication buildings.
  • Electrical substations and large-scale power infrastructure.
  • Gas turbines and battery energy storage.
  • Water treatment and recycling systems.
  • Industrial wastewater facilities.
  • Cooling, chemical distribution, and supporting utilities.

The scale of these requirements helps explain why establishing an advanced semiconductor factory can be considerably more difficult than assembling consumer electronics.

The building itself is only one component of the investment.

Why Tesla Shareholders Should Pay Attention

Terafab creates a fundamental question for investors in Tesla.

Could owning semiconductor manufacturing capacity become a durable competitive advantage?

Potentially.

If Tesla can manufacture large quantities of specialized AI chips economically, it may gain greater control over production costs, product development, and supply availability.

That could become valuable if Optimus robot production expands into the millions of units or autonomous-driving systems require significantly greater computing capacity.

But there is also another possibility.

Tesla could spend enormous amounts of capital building expertise that existing semiconductor companies already possess.

Chip manufacturing involves large fixed costs, continual equipment upgrades, and significant financial risk when production yields disappoint.

And the company’s current financial performance must support multiple other investment priorities.

According to Tesla’s third-quarter 2026 financial report, the company generated approximately $1.5 billion in quarterly free cash flow and ended the period with roughly $45.4 billion in cash, cash equivalents, and short-term investments.

Those are substantial resources.

Nevertheless, a multi-decade semiconductor project potentially requiring more than $100 billion represents a major capital-allocation decision.

During Tesla’s October 21 earnings call, Musk said the companies still need to determine how Terafab’s investment costs will be divided between Tesla and SpaceX.

That is an important unanswered question.

If Tesla contributes billions of dollars, shareholders should understand exactly which manufacturing assets, intellectual property rights, production capacity, and potential profits belong to Tesla.

Could a Tesla–SpaceX Merger Change the Economics?

The corporate picture has become more complicated.

During SpaceX’s October 20 earnings call, Musk confirmed that independent committees at Tesla and SpaceX are discussing a possible merger.

He identified Terafab and chip development among the potential advantages of combining the businesses.

However, no transaction terms had been agreed upon.

A merger is not guaranteed.

Until an agreement is announced and approved, it would be premature to assume that Tesla shareholders will own SpaceX’s Terafab-related assets—or that they will automatically benefit from all production at the shared project.

For long-term investors, ownership structure matters just as much as manufacturing capability.

The Real Challenge Is Not Building Chips. It’s Building Them Profitably.

There’s a reason companies such as TSMC, Samsung, and Intel have spent decades building their semiconductor businesses.

Manufacturing advanced chips is extraordinarily difficult.

A factory may successfully produce wafers without being competitive.

What ultimately matters is how many functioning chips it can produce, at what cost, with what reliability, and at what volume.

One important metric is manufacturing yield.

If a large percentage of chips on each wafer are defective, production costs can become prohibitive.

Another consideration is utilization.

Semiconductor factories contain expensive equipment that must operate efficiently to generate attractive returns.

If demand falls below expectations—or if a manufacturing process becomes obsolete—the financial consequences can be enormous.

Terafab also faces another strategic question.

Its stated long-term ambitions depend partly on massive demand for Optimus robots and orbital AI computing systems.

Those markets could expand dramatically.

But the scale, timing, and economic viability of that demand have not yet been established.

Terafab’s success will ultimately depend on producing valuable chips at competitive costs—not simply constructing the world’s largest semiconductor factory.

Why This Matters Beyond Tesla and SpaceX

Terafab illustrates a broader change taking place across the artificial intelligence economy.

AI is often discussed as a software revolution.

But behind every advanced AI system lies an enormous industrial supply chain.

Electricity must be generated.

Data centers must be constructed and cooled.

Semiconductors require specialized equipment, chemicals, minerals, and precision manufacturing.

That creates opportunities for companies in electrical engineering, industrial manufacturing, construction, automation, water treatment, and other essential industries.

Even North Carolina participates in this supply chain.

Near Spruce Pine, exceptionally pure quartz is extracted for use in specialized semiconductor manufacturing components.

As previously explored by Creative Loafing Charlotte in our analysis of the AI infrastructure economy, some of the most important businesses enabling artificial intelligence operate far from the consumer-facing software industry.

Terafab takes that idea to an extreme.

Tesla and SpaceX are betting that the physical infrastructure required to manufacture intelligence will become so economically and strategically important that building a vertically integrated semiconductor operation is worth the expense.

Whether that investment pays off remains an open question.

Five Milestones That Will Determine Terafab’s Future

The announcements are impressive, but the next two years should provide far more meaningful evidence.

1. Equipment commissioning in Austin. Does the research facility begin operating its semiconductor manufacturing systems during the first half of 2027?

2. A functioning test chip. Can Tesla produce working silicon by May or June 2027, as Musk hopes?

3. Repeatable advanced manufacturing. Can engineers progress from experimental chips to manufacturing processes with consistent performance and acceptable defect rates?

4. Production readiness at Grimes County. Are cleanrooms, equipment, electrical infrastructure, and supporting systems ready for commercial-scale operations in 2028?

5. Attractive manufacturing economics. Can Terafab eventually produce useful chips at costs that justify its immense capital investment?

The first two milestones are technical proof points.

The final three determine whether Terafab can become a successful industrial enterprise.

There is a significant difference between proving that a process is possible and proving that it creates economic value.

The Bottom Line: Terafab Is Real, but the Hardest Work Is Still Ahead

Tesla and SpaceX have made measurable progress.

Construction has started at the Austin research facility.

The primary production site has been announced, and earthwork is underway.

Intel has joined the initiative.

TSMC has acknowledged discussions.

Critical semiconductor equipment orders are progressing.

And the companies have disclosed ambitious dates for their first functional test chips and eventual production.

These developments distinguish Terafab from a concept presented only in promotional videos.

But they do not establish that the project has overcome the industry’s greatest technical challenges.

The first functioning chip will be an important achievement.

Commercial-scale production will be a much larger one.

Sustained, profitable manufacturing could be transformative.

Elon Musk has already demonstrated that seemingly impossible industrial objectives can sometimes be achieved through unconventional engineering and extraordinary capital commitments.

But semiconductor fabrication will test those capabilities in an industry where the smallest physical defects can determine whether billions of dollars produce profits or losses.

The most important date on Terafab’s calendar is now May–June 2027.

If Tesla can manufacture its first functioning test chip on schedule, it will demonstrate that the project is beginning to cross the boundary from construction into actual semiconductor production.

If it can then manufacture advanced chips reliably and competitively at scale, Terafab could become one of the most consequential industrial projects of the coming decade.

Until then, the most important distinction for investors, entrepreneurs, and the public is simple:

Terafab has demonstrated commitment. It has not yet demonstrated commercial semiconductor manufacturing success.

Sources and Further Reading

Editor’s note: This article distinguishes completed construction milestones, officially announced investments, management targets, and proposed future expansion phases. Manufacturing timelines, investment requirements, and production targets may change. This report is informational and does not constitute investment advice.

Leave a comment

Your email address will not be published. Required fields are marked *