NOW LET US – AI RAG SaaS Studio TP.HCM
NOW LET US
Digital Product Studio
Back to news
AI-FRONTIER...6 min read

Can Cursor Remain a Platform for OpenAI and Anthropic’s Models Inside SpaceX?

Share
NOW LET US Article – Can Cursor Remain a Platform for OpenAI and Anthropic’s Models Inside SpaceX?

SpaceX's $60 billion acquisition of Cursor raises critical questions about whether the popular AI coding tool can remain an open platform or if rivals like OpenAI and Anthropic will pull their models.

When SpaceX announced last month that it had agreed to acquire the popular AI coding startup Cursor for $60 billion, investors believed the deal would be a boon for both companies. Cursor would benefit from getting the computing resources of a major AI lab, which it could use to train its own models. In turn, SpaceX and Elon Musk would own one of the most popular AI developer tools on the market.

What was less clear was whether Cursor could remain an open platform after the deal, or if rival AI labs would continue letting it offer their models. Third-party AI models have historically played a critical role in Cursor’s business. While the company has started training its own AI models in recent years, it has always allowed users to choose from a variety of offerings from Anthropic, OpenAI, and other AI labs to power its coding assistant.

That strategy allowed Cursor to offer customers whichever model was the best, or cheapest, at a given moment. It also benefited Anthropic and OpenAI, which both count Cursor among their largest customers and feature the startup prominently in their marketing materials.

After SpaceX’s acquisition is finalized later this year, Cursor hopes to continue operating its AI coding product as a platform—serving models from Anthropic, OpenAI, and other AI labs alongside its own—according to people close to Cursor.

I have my doubts about how this will actually play out, but whether or not Cursor remains model agnostic is one of the biggest questions hanging over the AI industry.

Eno Reyes, the cofounder and chief technology officer of Factory, a smaller AI coding startup that competes with Cursor, says he’s not certain that SpaceX’s rivals will automatically cut Cursor off just because it will be owned by a competing AI lab. “I don't know if the decision is as black and white,” Reyes tells me. “It’s actually super unclear to us.”

Cursor declined to comment for this story. Anthropic, OpenAI, and SpaceX did not respond to requests for comment.

Making Frenemies

This is not the first time that Cursor’s relationship with OpenAI and Anthropic has been tested. Historically, Cursor complemented the AI labs by distributing their models through its coding platform. But now it has increasingly found itself in direct competition with them as OpenAI’s Codex and Anthropic’s Claude Code have become major lines of their respective businesses. The SpaceX acquisition will likely only intensify that rivalry.

SpaceX and Cursor can’t say a lot about how they’ll operate post-acquisition, in part, because the deal has not yet closed and remains subject to “requisite regulatory approvals,” according to documents SpaceX filed with the US Securities and Exchange Commission. But SpaceX is poised to get Cursor’s assets, customer contracts, and intellectual property—meaning that OpenAI and Anthropic will now have to do business with Musk if they want to reach Cursor’s users.

Once the acquisition is finalized, it’s possible SpaceX will decide it doesn’t want to send business towards Anthropic and OpenAI, two of its biggest competitors in the frontier AI development space. Anthropic and OpenAI may determine they are unwilling to sell their AI models through a product owned by Musk, who both companies’ CEOs, Dario Amodei and Sam Altman, have butted heads with in the past.

Historically, AI labs have not played nicely when it comes to selling AI models to one another. Last year, Anthropic was quick to cut off access to Windsurf after news broke that OpenAI was acquiring the AI coding startup (the deal ultimately didn’t pan out). Anthropic cofounder Jared Kaplan said at the time that it “would be odd to sell Claude to OpenAI.” In the months since, Anthropic has worked to limit OpenAI and SpaceX from using its Claude AI models.

But times may be changing. Anthropic recently struck a multi-billion dollar deal to buy computing resources from SpaceX, which suggests that Amodei and Musk may be willing to put aside their differences for the sake of beating their mutual enemy: OpenAI. That compute partnership may be reason enough for Anthropic to continue offering its AI models in Cursor.

OpenAI may find that it has different reasons to continue working with Cursor. The startup is a major partner of OpenAI, and the AI lab’s executives held preliminary discussions about acquiring it in the past. OpenAI’s startup fund was also one of the earliest investors in Cursor, participating in the company’s seed and Series A funding rounds. OpenAI’s startup fund is poised to see a significant return on its Cursor investment in the form of SpaceX stock as a result of the acquisition, according to people close to Cursor.

OpenAI says on its website that the company itself is not directly an investor in OpenAI’s startup fund, which was originally set up and managed by Altman. The startup fund receives investment from outside parties, such as Microsoft, as well as other OpenAI partners.

Is Independence Important?

Palantir CEO Alex Karp highlighted a broader concern I’ve been hearing from the AI industry in a viral CNBC appearance this week: Businesses are getting tired of being locked into the frontier AI labs, and want more options.

Reyes, the Factory CTO, says that “model independence”—the ability to avoid being tied to any one AI lab’s technology—is important to the Fortune 500 companies he speaks with because it offers them flexibility. Reyes believes this is one of the key advantages that independent AI coding startups like his have over the major AI labs. In the past, Cursor has highlighted its independence as an advantage as well.

However, there are significant benefits to working with an AI lab directly and being more than just a platform. Cursor CEO Michael Truell announced at its Compile conference last month that the startup is already partnering with SpaceX to train its next AI model, which will use ten to twenty times more computing power than the company could previously access. The hope is that will make the new model comparable to, or even better than, what OpenAI and Anthropic are offering. In a blog post from April, Cursor said that its lack of computing resources has been holding it back, and it now believes it can dramatically improve its models by relying on SpaceX’s data centers.

At the Compile conference, Truell added that Cursor is training its new AI model to be “intelligent beyond coding.” Over the last year, Cursor has started targeting other potential customer demographics beyond software engineers, shipping features catered toward people like graphic designers. After the acquisition closes, I wouldn’t be surprised if Cursor effectively becomes an enterprise AI arm of SpaceX.

Another factor to consider: Smaller AI coding startups are struggling to compete right now with the highly subsidized AI coding subscriptions that OpenAI and Anthropic offer developers. WIRED previously reported that OpenAI and Anthropic’s $200 monthly subscription plan can provide coders with well over $1000 of model usage. Now that Cursor is part of SpaceX, it may also be able to offer similarly aggressive pricing.

When I visited Cursor’s office a few months ago, shortly before news broke about the SpaceX acquisition, I argued that the startup’s main problem was that it didn’t have enough capital and computing power to achieve its lofty ambitions. I’d argue that Cursor is better off inside of SpaceX, even if it loses its relationship with OpenAI and Anthropic. But if Cursor can play nice and compete fiercely at the same time, then this might end up being one of the great acquisitions of the AI era.

© 2026 Now Let Us. All rights reserved.

Source: Wired AI

Advertisement
Ad slot ready: 5887729102

More in this category

NOW LET US Related – Teaching AI to run with the turbines

ai-frontier

Teaching AI to run with the turbines

Woodside Energy is moving beyond consumer AI hype, leveraging agentic AI and predictive models to optimize complex industrial workflows. By focusing on data governance and human-AI collaboration, the company showcases the future of the autonomous enterprise.

NOW LET US Related – OpenAI floats giving Trump administration 5 percent cut of AI boom

ai-frontier

OpenAI floats giving Trump administration 5 percent cut of AI boom

OpenAI has proposed giving the US government a 5 percent ownership stake, worth roughly $42.6 billion, to ease tensions with the Trump administration and avoid strict regulations.

NOW LET US Related – Meta Is Charging a Subscription for Smart Glasses Features. Welcome to the New Era of Consumer Tech

ai-frontier

Meta Is Charging a Subscription for Smart Glasses Features. Welcome to the New Era of Consumer Tech

Meta is introducing a subscription plan for its smart glasses to unlock advanced features like Conversation Focus. This move highlights a growing trend in consumer tech where hardware purchases are increasingly tied to monthly subscription fees.

NOW LET US Related – Goose, a New Gay Dating App, Appears to Be a Psyop

ai-frontier

Goose, a New Gay Dating App, Appears to Be a Psyop

Goose, a new dating app for gay men, is facing backlash after evidence emerged that it is using a network of AI-generated male influencers to deceptively promote the app through Instagram DMs and Close Friends Stories.

NOW LET US Related – The latest AI news we announced in June 2026

ai-frontier

The latest AI news we announced in June 2026

Google announced major AI updates in June 2026, including the local-running Gemma 4 12B model, Android 17 integration, and advanced tools for translation, education, and scientific research.

NOW LET US Related – You Can Now Sound the Alarm on AI Behaving Badly

ai-frontier

You Can Now Sound the Alarm on AI Behaving Badly

A group of AI researchers has launched FLARE-AI, a crowdsourced platform to report and track AI harms, aiming to bring transparency to a fragmented ecosystem.

EXPLORE TOPICS

Discover All Categories

Deep dive into the specific technology sectors that matter most to you.