Lovable Co-founder Anton: Creativity is the Only Moat for Products
"Vibe coding" representative startup Lovable's co-founder and CEO Anton Osika recently stated on social media that "creativity is becoming the only moat for building great products."
Osika, a Swedish entrepreneur, co-founded Lovable, an AI-based natural language programming tool aimed at enabling users without traditional programming backgrounds to generate usable software products directly through text descriptions. The company has been referred to by multiple media outlets as "one of the fastest-growing startups in history." According to previous reports, Lovable achieved an annual recurring revenue (ARR) of $10 million within 60 days of its establishment, with a team of only about 15 people.
Osika's remarks come at a time when AI-assisted programming tools are becoming widespread and the barriers to software development are significantly lowered. Products like Lovable and other "vibe coding" tools can now allow users to generate fully functional application prototypes within minutes to hours based solely on natural language descriptions. The proliferation of such tools has led to a continuous decline in the scarcity of "technical implementation capability" in product competition.
Osika emphasizes that as AI tools level the playing field in the execution phase of "bringing ideas to life," the key factor that will determine whether a product stands out will no longer be the engineering capabilities of the team, but rather the creativity to propose differentiated product ideas and insights into real user needs.
His viewpoint was publicly shared and discussed within 13 minutes, resonating with the ongoing discussions in Silicon Valley's startup community about the "shift of moats in the AI era." Related discussions have previously touched on various candidate factors for "moats," including distribution channels, branding, and data accumulation.
Source: Public Information
ABAB AI Insight
The definition of "moat" in the software industry is continuously evolving. In the 1990s to 2000s, companies like Microsoft and Oracle built their moats primarily on proprietary technology and high development barriers. In the 2010s, as open-source tools and cloud computing infrastructure became widespread, the technical implementation barriers significantly lowered, leading companies like Facebook and Airbnb to shift their moats towards network effects and user scale. Now, with AI programming tools like GitHub Copilot, Cursor, and Lovable making "writing code" extremely low-cost, discussions within the industry about whether "technical execution capability can still constitute a moat" are clearly heating up, aligning closely with Osika's remarks at this stage of the industry.
The continued investment in the "vibe coding" sector by venture capital over the past two years is itself a bet on the judgment that "execution barriers are disappearing, and creativity is becoming a scarce resource." AI programming tool startups, including Lovable, Replit, and Bolt, have received significant funding in the past couple of years, with capital more willing to pay a premium for teams that can demonstrate unique product judgment and user insights, rather than simply for the size of engineering teams. This contrasts sharply with the traditional SaaS funding logic that emphasizes evaluating team engineering capabilities and technical barriers.
This mirrors the historical shift in the photography industry from "film printing technology barriers" to "composition and narrative ability," and in the design industry from "typesetting software operation barriers" to "aesthetic judgment." Whenever a previously scarce execution skill is leveled by tool proliferation, the focus of industry competition systematically shifts from "who can do it" to "who knows what to do." The current AI programming tool industry is at a transitional inflection point from "proving AI can write code" to "discussing what companies can still compete on after AI writes code."
Essentially, this represents a "transfer of pricing power"—mechanistically, as engineering execution capabilities are commoditized by AI and marginal costs approach zero, the premium previously captured by the "technical execution" segment of the product value chain is shifting towards "creative judgment" and "demand insights," which are difficult for AI to replicate directly. This also suggests that in the future, the focus of organizational structure and talent competition may shift from expanding the size of engineering teams to competing for a small number of core talents with product intuition and creative judgment.
ABAB News · Cognitive Law
- Tools level execution power; creativity is the final differentiator.
- Those whose work can be easily replaced by AI will see their moats dry up first.
- More people can do the work, but knowing what to do becomes increasingly valuable.