The reported SpaceX Cognition acquisition discussions offer another glimpse into how aggressively Elon Musk is trying to build an AI business around SpaceX.
Bloomberg reported on August 19 that SpaceX had attempted to acquire Cognition, the company behind AI software engineering agent Devin. The report cited people familiar with the matter. Cognition CEO Scott Wu quickly rejected the account on X, saying the report was inaccurate and that Cognition is not for sale.
That distinction matters. At this point, the SpaceX Cognition acquisition story should be treated as a report of attempted discussions, not a completed deal or even an active acquisition process. Bloomberg reportedly said the acquisition talks are no longer active, while the companies may still be discussing a commercial relationship.
The broader story, however, is significant. SpaceX has already made a $60 billion deal for Cursor, acquired Musk’s xAI business, and publicly told investors that artificial intelligence could eventually represent the overwhelming majority of the company’s value.
What happened with the reported SpaceX Cognition acquisition
According to Bloomberg, SpaceX approached Cognition about a potential acquisition. Wu disputed that characterization shortly afterward, stating that Cognition is not for sale and that the two companies have not been in talks. Because the two accounts conflict, it would be premature to describe the situation as a confirmed acquisition attempt.
What is clear is that Cognition has become an increasingly valuable target in the AI software market.
The company raised more than $1 billion in May at a $25 billion pre-money valuation, according to TechCrunch. Cognition said at the time that enterprise use of Devin was expanding rapidly and reported an annualized revenue run rate of $492 million.
That makes the reported SpaceX Cognition acquisition particularly interesting. Cognition is no longer an early-stage startup that a major technology company could quietly absorb at a modest price. It is a large independent AI company with substantial enterprise ambitions.
Why Cognition would make sense for SpaceX
The most obvious strategic connection is coding. SpaceX already has Cursor under its umbrella following the $60 billion stock acquisition announced in June. Cursor was positioned as a key component of SpaceX’s effort to expand its AI capabilities and compete with major AI laboratories.
Cognition would potentially add something different. Its Devin product is designed to operate as an autonomous software engineering agent capable of taking on software development tasks rather than simply suggesting individual lines of code.
That distinction is important because the AI coding market is moving toward agents that can complete larger pieces of work with less direct human intervention.
For SpaceX, acquiring Cognition could therefore have provided another major software engineering platform, another enterprise customer base, and another team experienced in building AI agents for professional users.
But that does not mean SpaceX needs Cognition. Its acquisition of Cursor already gives the company a significant position in AI-assisted software development. Buying both companies could also create considerable integration challenges, particularly if their products, models, teams and customer strategies overlap.
Cursor has already changed SpaceX’s AI strategy.
The reported SpaceX Cognition acquisition cannot be understood separately from SpaceX’s deal for Cursor.
In June, SpaceX agreed to acquire Cursor in a transaction valued at $60 billion in stock. The deal followed SpaceX’s blockbuster IPO and came after the two companies had already established a development partnership. Cursor was not simply a convenient software purchase.
SpaceX investors had been presented with a huge opportunity in enterprise AI applications, while the company was also planning substantial investments in AI computing infrastructure. TechCrunch reported that SpaceX had identified a potential $22.7 trillion opportunity in enterprise applications and a $2.4 trillion AI infrastructure opportunity in its IPO materials.
That puts the SpaceX Cognition acquisition report into a much larger context. Musk appears to be pursuing several layers of the AI stack at once. There is the underlying AI model and research capability through xAI. There is AI infrastructure and computing capacity. Then there are applications that businesses and developers can actually pay for. Coding agents sit directly in that final category.
The economics explain why coding has become such a competitive part of the AI industry. Companies can sell coding agents directly to businesses and developers. Those customers have measurable workloads, recurring software budgets and an obvious reason to pay if the technology helps them ship software faster.
Cognition’s growth illustrates the opportunity. When it announced its May funding round, the company said its enterprise customer base included organizations such as Mercedes-Benz, NASA, Goldman Sachs and Santander. Cognition has also expanded beyond Devin itself.
In July, it acquired Poke, an AI assistant company, with the goal of bringing Poke’s interaction model and personality into Devin while giving Poke access to Cognition’s models and infrastructure.
That suggests Cognition is building toward a broader AI agent platform rather than remaining narrowly focused on code completion. For a company like SpaceX, that could be attractive.
The enterprise angle may be more important
The SpaceX Cognition acquisition story is not only about which company has the better coding model.
Enterprise distribution may be just as valuable. Large companies are increasingly experimenting with AI agents for software development, research, customer service, and internal operations. Once an AI product becomes embedded into those workflows, replacing it can become considerably harder.
Cognition’s enterprise relationships therefore represent a strategic asset in their own right. This is one reason SpaceX’s interest in AI coding makes sense even if the company can develop models internally. Building a capable model is only one part of creating a successful enterprise AI business.
You also need customers, distribution, product teams, security infrastructure, and an understanding of how businesses actually deploy AI.
The SpaceX Cognition acquisition report also needs to be viewed against Musk’s broader vision for the company. SpaceX completed its merger with xAI earlier this year, bringing Musk’s AI operation into the space company. SpaceX subsequently went public in June in what became one of the largest IPOs in history.
The company’s market value surged after the listing, briefly approaching $2.3 trillion on its first trading day. SpaceX has also said it plans to invest heavily in AI computing infrastructure.
That creates an unusual corporate strategy. SpaceX is simultaneously a launch company, satellite internet provider, AI infrastructure operator and AI software company.
Musk’s reported expectation that AI could eventually account for 99 percent of SpaceX’s value makes the software side particularly important.
The SpaceX Cognition acquisition report therefore looks less like an isolated piece of M&A speculation and more like another example of the company searching for ways to turn its AI ambitions into commercially valuable products.
Why Cognition may prefer to remain independent
There is another side to the story. Cognition has just demonstrated that it can raise enormous amounts of capital without selling itself. Its May financing valued the company at roughly $26 billion post-money, according to TechCrunch.
An independent company can also make decisions without having to fit into the priorities of a giant organization. That may be particularly valuable in AI, where product cycles are extremely short.
Cognition can focus on Devin, enterprise software development, and adjacent AI agents. It can choose its infrastructure partners and potentially work with multiple model providers rather than becoming tied to one corporate ecosystem.
If the reported SpaceX Cognition acquisition discussions really did occur, Cognition’s decision not to sell would therefore be understandable. The company may believe that its future value as an independent AI platform is greater than the immediate value of an acquisition offer.
SpaceX could still work with Cognition.
The most interesting possibility may be neither an acquisition nor a complete separation. Bloomberg reported that acquisition talks are no longer active but that the companies could still discuss working together, including the possibility of Cognition using SpaceX’s computing capacity.
That arrangement would make strategic sense. SpaceX could monetize infrastructure while Cognition gets access to substantial computing resources. Cognition would retain its independence while SpaceX gains another important AI customer and potential technology partner.
SpaceX is already positioning its computing infrastructure as a business that can serve AI companies before the company requires that capacity for its own operations. This model would allow SpaceX to benefit from Cognition without taking on the cost and complexity of an acquisition.
There is also an interesting cultural dimension to the SpaceX Cognition acquisition story. Cognition developed a reputation for an unusually intense working environment after acquiring the remaining assets of Windsurf in 2025. The company subsequently made layoffs and offered buyouts to Windsurf employees, according to TechCrunch.
Musk’s companies have similarly become associated with demanding work expectations. That does not mean the companies would naturally fit together. Corporate culture is more complicated than long hours. But the similarity is notable because successful acquisitions often depend as much on management philosophy as technology.
Cognition has its own momentum.
Perhaps the strongest argument against assuming a deal is imminent is Cognition’s recent trajectory. The company has continued making acquisitions, raising capital, and expanding Devin. Its $1 billion financing demonstrated that investors are willing to fund the company at a very high valuation.
Cognition also has an opportunity to position Devin as a broader workplace agent. If AI coding agents evolve into general-purpose software engineering systems that can plan projects, interact with development tools, test applications, and coordinate work, the market could become substantially larger than traditional coding assistance. That possibility gives Cognition a reason to keep building rather than selling.
The bigger lesson is that the AI race is shifting. The first phase was largely about building increasingly capable foundation models. The next phase is about turning those models into products that people and companies use every day.
Coding is one of the clearest examples because software development is structured, measurable, and highly compatible with AI agents.
SpaceX’s acquisition of Cursor shows that Musk understands the importance of that layer. The reported SpaceX Cognition acquisition attempt, if accurately characterized by Bloomberg, would suggest the company was considering doubling down on it. But Cognition’s denial is equally important. There is no confirmed acquisition of Cognition, and readers should not confuse a reported approach or discussion with a signed transaction.
What to watch next
The next development will probably be less dramatic than an acquisition announcement. Watch whether SpaceX and Cognition deepen their infrastructure relationship. Watch whether Devin gains closer access to SpaceX or xAI models. And watch whether SpaceX continues buying or partnering with companies that give it direct access to enterprise AI customers.
Cursor will also be an important test. SpaceX has already committed an enormous amount of capital to AI coding through the $60 billion Cursor transaction.
If that business grows rapidly inside SpaceX, the strategic case for acquiring another coding agent becomes weaker.
If Cursor struggles to deliver the enterprise scale SpaceX wants, Cognition or another AI software company could become more attractive.
For now, the SpaceX Cognition acquisition remains an unconfirmed acquisition story rather than a confirmed transaction. What is not speculative is the direction of SpaceX’s strategy.
Musk has put AI near the center of the company’s future, and software development is one of the most commercially promising areas in which that ambition can be tested.
According to the Tech Insider report, this story because the important development is not simply whether one acquisition happens. It is the increasingly aggressive battle among technology companies to control the AI software layer.
Cognition’s refusal to sell, if its CEO’s statement is taken at face value, may ultimately make the company more interesting, not less. An independent Devin gives SpaceX a potential partner, a potential infrastructure customer and, perhaps most importantly, another benchmark against which its own AI strategy can be measured.
What Happened Next
The reported SpaceX Cognition acquisition is not a completed deal, and Cognition’s CEO has explicitly disputed the report. Still, the episode highlights something important about the AI market in 2026.
SpaceX is no longer treating AI as a side project. Through xAI, Cursor, computing infrastructure, and its broader enterprise ambitions, the company is attempting to build a serious AI business around its existing space and connectivity operations.
Cognition is exactly the kind of company that fits that strategy. Whether the relationship becomes an acquisition, a commercial partnership, or nothing more remains uncertain. For now, the clearest signal is that AI coding has become strategically important enough that one of the world’s most valuable companies is looking closely at the companies building it.
