SpaceX has officially completed its acquisition of Cursor, the AI-powered code editor that became a favorite among developers over the past two years. The deal, which was first reported in June, marks another major consolidation in the AI coding tools market.

Financial terms weren’t disclosed, but the acquisition gives SpaceX one of the most popular AI coding assistants on the market. Cursor built its reputation on a cleaner interface and better context awareness than GitHub Copilot, attracting users who wanted something that felt less like autocomplete and more like pair programming.

For SpaceX, the move makes sense. The company has been public about wanting to bring more software development in-house, and owning the tools its engineers use daily fits that strategy. It’s also a signal about where Elon Musk sees value in AI: not in chatbots or image generators, but in tools that make his existing operations faster and cheaper.

The Cursor team will continue operating independently for now, according to a brief statement from SpaceX. That’s standard language for these deals, though the real test will be whether the product stays available to outside developers or eventually becomes SpaceX-only.

This isn’t the first time a major tech company has acquired an AI coding startup. Microsoft bought GitHub (which runs Copilot) back in 2018 for $7.5 billion. Google has been building its own tools internally. The pattern is clear: companies that employ thousands of software engineers want to own the AI that makes those engineers more productive.

Amazon’s Twitch problem

While SpaceX was closing deals, Amazon was dealing with blowback over how it uses content from Twitch, the streaming platform it owns.

Twitch quietly updated its terms of service to allow Amazon to use streamer content for training AI models. The company did include an opt-out option, but thousands of users are asking why their videos, commentary, and creative work are being scraped for AI training in the first place.

The backlash highlights a growing tension in the creator economy. Platforms built their businesses by hosting user content, and now they’re treating that same content as raw material for AI development. Twitch streamers spend hours building audiences and producing content. Finding out that work is being fed into Amazon’s AI systems without explicit consent didn’t go over well.

Amazon’s defense is that it gave users an opt-out. That’s technically true, but it misses the point. The default should be opt-in, especially when you’re talking about using someone’s creative work to build a commercial product.

This isn’t just a Twitch issue. Every platform with user-generated content is facing the same question: do they have the right to use that content for AI training? Reddit made a similar move and faced similar criticism. Twitter (now X) did the same. The pattern is predictable: announce the policy change, add an opt-out buried in settings, wait for the anger to fade.

What’s different about Twitch is the nature of the content. Streamers aren’t just posting text or photos. They’re often on camera, talking for hours, building parasocial relationships with their audiences. Using that material to train AI models feels more invasive than scraping forum posts.

The business logic is obvious. Amazon wants to train AI models, and it owns a platform with millions of hours of video content. But the ethical logic is messier, and the company hasn’t done a great job explaining why this is good for creators rather than just good for Amazon.

For now, streamers who care about this need to manually opt out. How many will actually do that remains to be seen. But the fact that Amazon structured this as opt-out rather than opt-in tells you everything about how it views the balance of power between platform and creator.