Midjourney has never been a company that stayed in one lane for long. It started as an image generator that made “AI art” feel immediate and playful—something you could try in minutes, share in group chats, and iterate on with the kind of obsessive curiosity that usually belongs to hobbyists. But over time, the company’s public footprint has suggested a broader ambition: not just making outputs, but building systems that can be repurposed across domains, from creative tools to high-stakes applications.
Now Midjourney is taking another step that looks, at first glance, almost comically far from where it began. The company has acquired Co-Star, a personalized astrology app that delivers daily horoscopes and compatibility insights. The move is notable not because astrology is new to AI—there are plenty of chatbots and content generators that can produce horoscope-style text—but because Midjourney is signaling that it wants to own a consumer experience where personalization, retention, and daily habit formation matter as much as the model behind the scenes.
The acquisition was announced this week, following earlier reporting by Bloomberg. Midjourney says it has acquired Co-Star, and while the deal terms were not disclosed, the timing suggests the transaction closed earlier this spring. For users, the most immediate question is simple: will Co-Star still feel like Co-Star? For the industry, the question is more revealing: why would an AI company known for creative generation want to buy a mainstream astrology product?
To understand the logic, it helps to look at what Co-Star actually does—and how it frames its approach. Co-Star is free to use and centers on daily horoscopes. It also offers compatibility features, letting users explore how they might align with friends or partners. In other words, it’s not just a content feed. It’s a relationship tool, a ritual, and a personalization engine wrapped in a familiar cultural format.
Co-Star’s own FAQ describes a blend of human insight, data from NASA, and AI to deliver individualized guidance each day. That combination matters. Astrology apps often rely on static interpretations—“your sign means X”—but Co-Star’s pitch is closer to “your chart and your context produce something tailored.” Even if you don’t believe in astrology, the product mechanics are recognizable: users return because the app feels like it knows them, and because the daily update creates a reason to check in.
That’s where Midjourney’s acquisition becomes less surprising. The company has already demonstrated that it can move beyond a single output type. If you can build a system that produces compelling images, you can also build systems that produce compelling narratives, explanations, and recommendations. The hard part isn’t generating text; the hard part is making it feel personal, consistent, and worth coming back to.
Astrology is an unusually effective testbed for those challenges. It’s inherently interpretive, which means the “right answer” isn’t objective. That gives product teams room to tune tone, pacing, and perceived relevance without needing clinical validation or strict factual correctness in the way many other domains require. At the same time, astrology is emotionally sticky. People don’t just read horoscopes; they use them to make sense of their day, their relationships, and their identity. That emotional stickiness is exactly what consumer AI products chase when they’re trying to move from novelty to habit.
Co-Star’s integration of “human insight” alongside AI is also a clue about how Midjourney may think about personalization. Many AI startups treat the model as the whole product. But in practice, successful consumer experiences often depend on a layered system: curated frameworks, editorial judgment, guardrails, and feedback loops that shape what the model is allowed to say and how it should say it. Astrology is a domain where those layers can be especially important, because the app must balance specificity with reassurance, and it must avoid sounding generic while still staying within a recognizable style.
There’s also the question of data. Co-Star references NASA data in its FAQ, implying that it uses astronomical inputs—likely related to planetary positions and timing—to ground its interpretations. That’s a different kind of “data pipeline” than the one used for image generation, but it’s still a pipeline. It’s still about turning raw inputs into user-facing meaning. Midjourney buying Co-Star could be interpreted as acquiring not only an app, but a working personalization workflow that already connects external data sources to daily user experiences.
For Midjourney, the acquisition may also represent a strategic shift in how it approaches product development. Image generation is powerful, but it’s not naturally a daily habit for most users. You can generate images whenever inspiration strikes, but the product doesn’t automatically create a reason to return tomorrow. Astrology does. A daily horoscope is a built-in cadence. Compatibility checks create social prompts. The app becomes part of routines.
That routine is valuable for any AI company because it changes the economics of engagement. Instead of relying on viral sharing or occasional experimentation, you can build a stable user base that returns predictably. And once you have predictable return behavior, you can improve personalization over time—whether through user feedback, interaction patterns, or iterative refinement of the content generation process.
Of course, there’s a tension here. Astrology is not universally accepted, and some users approach it as entertainment. Others treat it as guidance. An AI-powered version of astrology can intensify both reactions. If the app feels too confident, it can alienate skeptics. If it feels too vague, it won’t satisfy believers. The product challenge is to calibrate certainty and tone so that the experience feels meaningful without becoming manipulative.
This is where Midjourney’s track record could matter. The company has been expanding into more serious applications, including medical AI work reported in recent coverage. That doesn’t mean Co-Star will become a medical tool, but it does suggest Midjourney has experience thinking about reliability, evaluation, and the consequences of getting things wrong. Consumer astrology is not a clinical domain, but it is a domain where users can make decisions based on what they read. Even if the app is framed as guidance rather than truth, the psychological impact is real.
So what might change after the acquisition? In the short term, likely not much on the surface. Co-Star is currently free, and the acquisition terms weren’t disclosed. But integration work almost always affects the underlying system first: how content is generated, how personalization is computed, how the app handles edge cases, and how it responds to user interactions. Users may notice improvements in specificity, smoother language, or better alignment between what the app says and what users do in the app.
One unique angle to watch is whether Midjourney brings its own “style” to Co-Star’s outputs. Midjourney is known for distinctive creative sensibilities. If that sensibility translates into astrology writing, users could see a shift in voice—more vivid, more poetic, or more structured. Alternatively, Midjourney may choose to preserve Co-Star’s existing tone to avoid breaking trust. In consumer products, trust is fragile. A sudden change in voice can feel like a bait-and-switch even if the underlying personalization improves.
Another area to watch is compatibility features. Compatibility is one of the most socially shareable parts of astrology apps, and it’s also one of the easiest places for AI to overreach. If the app generates overly deterministic claims about relationships, it can trigger backlash. If it’s too cautious, it loses its appeal. The best compatibility experiences tend to frame insights as possibilities rather than verdicts. Midjourney’s integration could either refine that framing—or risk making it too intense.
Then there’s the question of “human insight.” Co-Star’s FAQ indicates that human input plays a role. That could mean editorial oversight, curated interpretation templates, or a hybrid system where AI drafts and humans refine. If Midjourney leans heavily into automation, the app could become cheaper to run but potentially less nuanced. If Midjourney keeps the human layer intact, it may preserve quality while still benefiting from AI improvements.
The acquisition also raises a broader industry question: what does it mean when AI companies buy consumer apps that aren’t obviously “AI-native”?
In many cases, AI startups build tools that developers integrate into their own products. But consumer AI is different. It requires a complete experience: onboarding, retention loops, content cadence, user trust, and a clear value proposition that survives beyond the novelty phase. Buying an existing app can shortcut that work. Co-Star already has a user base, a daily ritual, and a content engine. Midjourney isn’t starting from scratch; it’s acquiring a ready-made distribution channel and a proven engagement model.
This is part of a larger pattern in tech: when AI models get commoditized, the differentiator shifts toward product design and user experience. Models can be swapped. Interfaces and habits are harder to replicate. A company that owns a consumer ritual has leverage. It can experiment with new model versions, adjust personalization strategies, and iterate on the experience without asking users to learn a new tool every time.
Astrology, again, is a particularly interesting ritual because it’s already culturally embedded. People know what a horoscope is. They know what compatibility means. That reduces friction. The app doesn’t need to convince users that the concept is legitimate; it needs to convince them that the experience is personally relevant and emotionally satisfying.
Midjourney’s move also hints at where AI product teams may be heading next: toward “meaningful personalization” rather than raw generation. The future of consumer AI may not be about producing more content—it may be about producing the right kind of content at the right time, in the right tone, for the right person. Daily horoscopes are a perfect example of timed personalization. Compatibility checks are a perfect example of contextual personalization tied to relationships.
And there’s a final layer to consider: the business model. Co-Star is free, at least currently. Many free apps monetize through subscriptions, ads, or premium features. If Midjourney sees Co-Star as a platform, it may eventually introduce paid tiers—perhaps deeper compatibility analysis, more frequent