Incomputer programming,vibe coding is anAI-assisted software development practice. It is achatbot-based approach to creatingsoftware where the developer describes a project or task to alarge language model (LLM), which generatessource code based on theprompt. The term was introduced by computer scientistAndrej Karpathy, a co-founder ofOpenAI and former AI leader atTesla, in February 2025. According to Karpathy, vibe coding typically involves accepting AI-generated code without closely reviewing its internal structure, instead relying on results and follow-up prompts to guide changes.[1][2]
Merriam-Webster listed it as a "slang & trending" term in March 2025.[3] It was named theCollins English DictionaryWord of the Year for 2025.[4][5]
Advocates of vibe coding say that it allows evenamateur programmers to produce software without the extensive training and skills required forsoftware engineering.[6][7] Critics point out a lack of accountability, maintainability, and the increased risk of introducingsecurity vulnerabilities in the resulting software.[1][7]
The concept refers to a coding approach that relies on LLMs, allowing programmers to generate working code by providingnatural language descriptions rather than manually writing it.[1][2][7]
Karpathy described it as "fully giv[ing] in to the vibes, embrac[ing] exponentials, and forget[ting] that the code even exists."[8] He used the method to build prototypes like MenuGen, letting LLMs generate all code, while he provided goals, examples, and feedback via natural language instructions.[9] The programmer shifts from manual coding to guiding, testing, and giving feedback about the AI-generatedsource code.[1][2][10]
The concept of vibe coding elaborates on Karpathy's claim from 2023 that "the hottest newprogramming language is English", meaning that the capabilities of LLMs were such that humans would no longer need to learn specific programming languages to command computers.[11]
A key part of the definition of vibe coding is that the user accepts AI-generated code without fully understanding it.[1] ProgrammerSimon Willison said: "If an LLM wrote every line of your code, but you've reviewed, tested, and understood it all, that's not vibe coding in my book—that's using an LLM as a typing assistant."[1]
In February 2025,New York Times journalistKevin Roose, who is not a professional coder, experimented with vibe coding to create several small-scale applications. He described these as "software for one" due to the ability to personalize the software. However, Roose also stated that the results are often limited and prone to errors.[10][11] In one case, the AI-generated code fabricated fake reviews for ane-commerce site.[10]
In response to Roose, cognitive scientistGary Marcus said that the algorithm that generated Roose's LunchBox Buddy app had presumably been trained on existing code for similar tasks. Marcus said that Roose's enthusiasm stemmed from reproduction, not originality.[11]
In March 2025,Y Combinator reported that 25% ofstartup companies in its Winter 2025 batch had codebases that were 95% AI-generated, reflecting a shift toward AI-assisted development within newer startups.[12] The question asked was about AI-generated code in general, and not specifically about vibed code.
Inspired by "vibe coding",The Economist suggested the term "vibe valuation" to describe the very large valuations of AI startups byventure capital firms that ignore accepted metrics such asannual recurring revenue.[13]
In July 2025,The Wall Street Journal reported that vibe coding was being adopted by professional software engineers for commercial use cases.[14]
In July 2025, SaaStr founder documented his negative experiences with vibe coding:Replit's AI agent deleted a database despite explicit instructions not to make any changes.[15][16]
In September 2025,Fast Company reported that the "vibe coding hangover" is upon us, with senior software engineers citing "development hell" when working with AI-generated code.[17]
It was reported in January 2026 thatLinus Torvalds had made use ofGoogle Antigravity to vibe code a tool component of his AudioNoise random digital audio effects generator. Torvalds explained in the project'sREADME file that "thePython visualizer tool has been basically written by vibe-coding."[18][19]
Andrew Ng has taken issue with the term, saying that it misleads people into assuming that software engineers just "go with the vibes" when using AI tools to create applications.[20]
Vibe coding has raised concerns about understanding and accountability. Developers may use AI-generated code without fully comprehending its functionality, leading to undetected bugs, errors, orsecurity vulnerabilities.[21] While this approach may be suitable forprototyping or "throwaway weekend projects" as Karpathy originally envisioned, it is considered by some experts to pose risks in professional settings, where a deep understanding of the code is crucial fordebugging, maintenance, andsecurity.Ars Technica cites Simon Willison, who stated: "Vibe coding your way to a production codebase is clearly risky. Most of the work we do as software engineers involves evolving existing systems, where the quality and understandability of the underlying code is crucial."[1]
In May 2025,Lovable, a Swedish vibe coding app, was reported to have security vulnerabilities in the code it generated, with 170 out of 1,645 Lovable-createdweb applications having an issue that would allow personal information to be accessed by anyone.[22][23] In December 2025, computer security researcher Etizaz Mohsin discovered a security flaw in the Orchids vibe coding platform, which he demonstrated to aBBC News reporter in February 2026.[24]
A December 2025 analysis by CodeRabbit of 470 open-source GitHub pull requests found that code that was co-authored by generative AI contained approximately 1.7 times more "major" issues compared to human-written code. The study revealed that AI co-authored code showed elevated rates of logic errors, including incorrect dependencies, flawed control flow, and misconfigurations (75% more common), security vulnerabilities (2.74x higher). Additionally, they also reported high code readability issues, including formatting errors and naming inconsistencies.[25][26]
Vibe coding has the potential of making code harder to maintain in the longer term and leading totechnical debt.
In early 2025, GitClear published the results of a longitudinal analysis of 211 million lines of code changes from 2020-2024. They found that the volume ofcode refactoring dropped from 25% of changed lines in 2021 to under 10% by 2024,code duplication increased approximately four times in volume, copy-pasted code exceeded moved code for the first time in two decades, and code churn (prematurely merged code getting rewritten shortly after merging) nearly doubled.[27][26]
Generative AI is highly capable of handling simple tasks like basic algorithms. However, such systems struggle with more novel, complex coding problems like projects involving multiple files, poorly documented libraries, or safety-critical code.[28]
In July 2025, METR, an organization that evaluatesfrontier models, ran arandomized controlled trial to understand developer productivity involving generative AI programming tools available in early 2025. They found that experienced open-source developers were 19% slower when using AI coding tools, despite predicting they would be 24% faster and still believing afterward they had been 20% faster.[29][26]
LLMs generate code dynamically, and the structure of such code may be subject to variation.[30] In addition, since the developer did not write the code, the developer may struggle to understand its syntax and concepts.[28]
In January 2026, a paper authored by experts from several universities titled "Vibe Coding Kills Open Source"[31] argued that vibe coding has negative impact on theopen-source software ecosystem. The authors say that increased vibe coding reduces user engagement with open-source maintainers, which has hidden costs for said maintainers. Speaking withThe Register about their paper, the authors argued:[32]
"Vibe coding raises productivity by lowering the cost of using and building on existing code, but it also weakens the user engagement through which many maintainers earn returns," the authors argue. "When OSS is monetized only through direct user engagement, greater adoption of vibe coding lowers entry and sharing, reduces the availability and quality of OSS, and reduces welfare despite higher productivity."
They added that revenue is not the only thing that may be affected by this trend, as open-source software maintainers traditionally also get non-tangible benefits from their work, such as community recognition, reputation, and job prospects.
Maya Posch, explaining the paper's claims onHackaday, expanded on the explanation. She pointed out that the mechanism for vibe coding lowering harmony with open-source projects is the homogenization of software development; language models will gravitate towards large and established libraries that appear frequently in their training dataset, removing the organic selection process of libraries and tooling and making it harder for newer open-source tools to get noticed. She also pointed out that language models will not submit useful bug reports to the maintainers, or be aware of potential issues.[33]
The technique, enabled by large language models (LLMs) from companies like OpenAI and Anthropic, has attracted attention for potentially lowering the barrier to entry for software creation. But questions remain about whether the approach can reliably produce code suitable for real-world applications, even as tools like Cursor Composer, GitHub Copilot, and Replit Agent make the process increasingly accessible to non-programmers.
Karpathy's "vibe coding" is a recognition of how sophisticated AI systems have evolved. In describing on X (formerly Twitter), he added that LLMs, like the Cursor Composer with Sonnet, are advancing to a degree that nearly eliminates the use of traditional coding mechanisms. Describing his own experience, Karpathy explained how he converses with AI tools almost in a passive manner—merely talking to them and having the AI handle the rest. This method eliminates manually typing code as well as keeping track of all the minute information in the program.
Vibe coding (also written as vibecoding) (Vibecode/Vibecoder) is a recently-coined term for the practice of writing code, making web pages, or creating apps, by just telling an AI program what you want, and letting it create the product for you. In vibe coding the coder does not need to understand how or why the code works, and often will have to accept that a certain number of bugs and glitches will be present. The verb form of the word is vibe code.
Vibecoding, a term that was popularized by the A.I. researcher Andrej Karpathy, is useful shorthand for the way that today's A.I. tools allow even nontechnical hobbyists to build fully functioning apps and websites, just by typing prompts into a text box. You don't have to know how to code to vibecode — just having an idea, and a little patience, is usually enough. "It's not really coding," Mr. Karpathy wrote this month. "I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works."
| Concepts |
| ||||||
|---|---|---|---|---|---|---|---|
| Applications |
| ||||||
| Implementations |
| ||||||
| People |
| ||||||
| Architectures |
| ||||||
| Political | |||||||
| Social and economic | |||||||
