Key Takeaways
- Lovable reached $100M ARR in 8 months, the fastest any software company has ever achieved that milestone — without spending a dollar on paid acquisition.
- Cursor (Anysphere) crossed $2B in annualized revenue by March 2026 and raised at a $29.3B valuation, doubling its ARR in just three months.
- Replit tripled its valuation to $9B in six months after raising a $400M Series D, with $240M in 2025 revenue and a target of $1B ARR by end of 2026.
- Security researchers found more than 10% of Lovable-generated apps shipped with user data exposed, raising questions about whether growth can be sustained without addressing code quality.
What Happened
In February 2025, former OpenAI co-founder and Tesla AI director Andrej Karpathy posted a description of a new way to build software. He called it vibe coding: “There’s a new kind of coding I call ‘vibe coding’, where you fully give in to the vibes, embrace exponentials, and forget that the code even exists,” he wrote on X. The concept — describe what you want, let the AI build it, accept everything — spread fast enough that Merriam-Webster listed it as a trending term by March 2025 and Collins English Dictionary named it Word of the Year.
The companies that had been building tools for exactly this workflow were already generating extraordinary revenue. Lovable, a Swedish startup led by CEO Anton Osika, announced in July 2025 that it had crossed $100M in annual recurring revenue just 8 months after launch — faster than OpenAI, Cursor, Wiz, and every other software company in recorded history, according to TechCrunch. By the time Lovable announced a $330M raise at a $6.6B valuation in December 2025, ARR had doubled to $200M.
Cursor, the AI-first code editor built by Anysphere, followed its own steep curve. The company hit $500M ARR in May 2025, $1B ARR by October, and crossed $2B in annualized revenue by March 2026, per TechCrunch. It raised a $2.3B Series D at a $29.3B valuation in November 2025. For context: Slack took five years to reach $1B ARR. Zoom took nine.
Replit, the browser-based coding environment that pivoted hard into AI-assisted app creation, raised a $400M Series D at a $9B valuation in March 2026 — triple its $3B valuation from just six months earlier. The company reported $240M in 2025 revenue and more than 150,000 paying customers.
Why It Matters
The revenue numbers reflect a structural shift in who can build software. Traditional software development requires years of training. Vibe coding tools target the gap between a software idea and the ability to execute it — a gap that includes every entrepreneur, product manager, designer, and domain expert who has never written a line of code.
Anton Osika described the scale of that gap when Lovable hit $10M ARR in 60 days with 15 employees: the company had zero paid acquisition spend and 2.3 million active users. “It’s pretty fun being the frontman of the world’s fastest-growing company,” Osika wrote on LinkedIn. At $100M ARR, the company had 45 employees — a revenue-per-employee ratio of $2.2M.
Anysphere CEO Michael Truell attributed Cursor’s growth to value delivery for professional developers. “A lot of the excitement comes from the value that this tech is giving to developers,” he told Bloomberg. Enterprise customers now account for approximately 60% of Cursor’s revenue, a shift from its early developer-focused positioning.
Replit CEO Amjad Masad frames the opportunity in broader terms. “We think that code has become the universal language for knowledge work, and we want to make it so that anyone can build applications,” he said in Replit’s funding announcement.
Technical Details
The three companies occupy distinct but overlapping segments of the market.
Lovable targets non-developers who want full web applications. Users describe what they want in natural language; Lovable generates a deployable React app with a Supabase backend. The platform is optimized for speed and accessibility, not code ownership or extensibility.
Cursor targets professional developers. It is a fork of VS Code with deep model integration — supporting Claude, GPT-4o, and Gemini — that accelerates rather than replaces the developer. Features like Composer and multi-file editing make it a force multiplier for engineers who already code, not a replacement for coding knowledge.
Replit sits between both segments. Its browser-based IDE lets beginners start without setup while its Replit Agent handles autonomous multi-step tasks. The March 2026 launch of Agent 4 introduced a visual canvas interface where users can modify mockups and collaborate in real time.
The vibe coding market is projected to grow from approximately $3B in 2025 to $12.3B by 2027, at a 38% CAGR.
Who’s Affected
Professional developers are the most immediate group affected, with 92% of US developers now using AI coding tools daily, according to Second Talent’s 2026 vibe coding statistics. GitHub reports 46% of all new code is now AI-generated.
Non-technical founders and small businesses represent the growth frontier. Lovable’s 10 million projects created — with 100,000 new projects per day — suggests demand from people who previously had no path to software ownership.
Security professionals have flagged a more immediate risk. A 2025 analysis of 1,645 Lovable-created applications found that 170 had vulnerabilities exposing personal user data to anyone, according to The New Stack. A December 2025 CodeRabbit study of 470 open-source GitHub pull requests found AI co-authored code contained 2.74 times more security vulnerabilities than human-written code.
What’s Next
The bubble question centers on whether revenue growth reflects durable value or a novelty cycle. A randomized controlled trial by METR in July 2025 found that experienced open-source developers were 19% slower when using AI coding tools, despite predicting they would be 24% faster and believing afterward that they had been 20% faster. Code churn rates were 41% higher in AI-assisted workflows.
The optimistic case rests on the fact that vibe coding tools are not primarily competing with professional developers for complex systems work. They are unlocking a population that previously built nothing. If even a fraction of the people with software ideas but no coding ability convert to paying users, the TAM is enormous.
Replit targets $1B ARR by end of 2026. Lovable is already above $200M and has a $6.6B valuation to justify. Cursor’s enterprise pivot means its trajectory is tied to whether large organizations maintain or expand AI tool budgets as the productivity evidence matures.
Concrete next steps for anyone tracking this space:
- If evaluating vibe coding tools for a team, run a time-controlled pilot and measure output quality and churn rate, not just perceived speed.
- If building on any of these platforms, audit generated applications for exposed data endpoints before launch — the Lovable vulnerability findings suggest this step is routinely skipped.
- If investing or competing in this space, watch Replit’s $1B ARR target for 2026 as the clearest near-term test of whether the consumer-plus-enterprise model holds at scale.
