AI is influencing programming language use, but there is no single industry-wide ranking or evidence that AI alone caused recent shifts. GitHub’s rankings depend on what activity it counts: Python led its overall contribution measure in 2024, JavaScript still led code pushes, and TypeScript became GitHub’s most-used language in August 2025. Meanwhile, Stack Overflow respondents reported rising Python adoption. These are different measures, not competing answers to one universal question.
What the recent language rankings actually show
Language rankings depend on the data being counted. Contributions, code pushes, repository contents and survey responses describe different kinds of activity. GitHub’s platform trends reveal what happens on GitHub; they are not a census of all software development.
As an Amazon Associate I earn from qualifying purchases.
| Source and period | Measure | What it reported |
|---|---|---|
| GitHub, 2024 Octoverse | Overall contribution activity | Python overtook JavaScript as GitHub’s most-used language by this measure. JavaScript remained first for code pushes alone. |
| GitHub, August 2025 | GitHub language ranking | TypeScript became GitHub’s most-used language, overtaking Python and JavaScript. |
| Stack Overflow, 2025 survey | Respondents’ reported language adoption | Python adoption increased by seven percentage points from 2024 to 2025 among survey respondents. |
GitHub’s 2024 report connected Python’s rise with machine learning, data science, scientific computing and more participation from STEM communities. It also recorded a 92% increase in Jupyter Notebook use in its 2024 measurement. In GitHub’s August 2025 analysis, TypeScript had added more than one million contributors over the prior year; Python added roughly 850,000 contributors, a 48.78% year-over-year increase, and JavaScript roughly 427,000, a 24.79% increase. These are GitHub contributor counts under its methodology, not estimates of all developers worldwide.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsWhich languages are developers using for AI?
Python remains prominent in AI-focused work
GitHub reported that nearly half of new AI projects on its platform were primarily built in Python as of August 2025. That finding concerns AI-project activity on GitHub, not the share of all software written in Python. Separately, Stack Overflow’s 2025 survey found a seven-percentage-point increase in Python adoption among respondents from the prior year. Its survey asks: “Which programming, scripting, and markup languages have you done extensive development work in over the past year, and which do you want to work in over the next year?” Survey adoption and GitHub project activity are useful context, but they should not be treated as the same metric.
TypeScript is gaining ground in broader platform activity
TypeScript’s August 2025 position on GitHub reflects the platform’s overall language ranking, not a finding that it has displaced Python for AI projects. GitHub’s analysis argues that explicit types can help developers spot mismatches in AI-generated code. That is a plausible workflow advantage and GitHub’s interpretation of the trend; it is not proof that AI caused TypeScript’s rise.
Why AI may influence language choice
Models may be more familiar with widely used languages
Languages with extensive public code and documentation may give models more examples from which to generate familiar patterns. GitHub’s 2025 discussion describes popular languages including TypeScript, Python, Java and Go as areas where models are strong. Idan Gazit, head of GitHub Next, summarized the training-data argument this way: “If the model has seen a trillion examples of TypeScript and only thousands of Haskell, it’s just going to be better at TypeScript.” This is an explanation of incentives, not an independently measured cross-language benchmark.
Rank #2
Types and tooling can make review more concrete
When generated code has to fit existing interfaces, a type system can expose some mismatches before runtime. Tests, linters and other tooling can add further checks. These tools do not establish that a language is inherently safer or that generated output is correct; they make certain errors easier to detect within a project’s conventions.
Team experience and the existing ecosystem still matter
AI assistance is one factor among many. A team’s expertise, the application’s requirements, libraries, deployment environment and maintenance expectations may outweigh any difference in model familiarity. Switching languages solely because a model produces plausible snippets can create costs in integration, review and long-term support.
AI adoption does not mean prompt-only programming
In a GitHub survey of 2,000 respondents across the United States, Brazil, Germany and India, more than 97% said they had used AI coding tools at some point. The result measures whether respondents had ever used such tools, not how often they use them or whether AI changed their language choices. Participants described benefits that included faster adoption of programming languages.
Professional use remains more conventional than the phrase “vibe coding” might suggest. Stack Overflow’s 2025 survey found that 72% of respondents said vibe coding was not part of their professional development work, and another 5% emphatically said it was not part of their workflow. AI-assisted programming can fit into normal development—writing or explaining code, generating tests, or helping with unfamiliar syntax—without replacing design, review and debugging.
Generated code still needs review and debugging
In Stack Overflow’s 2025 survey, favorable sentiment toward AI tools was 60%, down from more than 70% in both 2023 and 2024. Among respondents who answered the survey’s frustration question, 66% said they encountered AI solutions that were almost right but not quite, and 45% said debugging AI-generated code took more time. These responses are self-reported, but they underline a practical point: code generation does not remove the need to verify behavior, test edge cases and maintain the result.
Recommended Free Tools
- Check generated code against the project’s interfaces, requirements and security practices.
- Run relevant tests and add coverage for edge cases the prompt may have omitted.
- Review dependencies, error handling and assumptions before merging or deploying.
- Prefer code your team can understand and support over output that merely looks plausible.
What estimates of AI-written code can—and cannot—tell us
A 2025 working paper, “Who is using AI to code? Global diffusion and impact of generative AI”, analyzed 80 million GitHub commits from 2018 through 2024 and used a classifier to estimate AI-generated Python functions. For December 2024, it estimated the AI-written share of Python functions at 30.1% for U.S. contributors, 24.3% for Germany, 23.2% for France, 21.6% for India, 15.4% for Russia and 11.7% for China. These figures apply to Python functions in the study’s commit sample and depend on its classifier; they are not shares of all code, all languages or all programming work in those countries.
Best Value
The paper’s within-developer fixed-effects model associated moving to 30% AI use with 2.4% more quarterly commits. That is a model estimate in the paper’s data and design, not a general productivity promise or proof that AI caused a corresponding increase in output.
How to choose a language when AI is part of the workflow
Use AI support as a practical consideration, not a substitute for choosing a language that fits the work. Compare options against the project’s constraints:
- Workload: Match the language and its ecosystem to the application—such as data analysis, web development or backend services.
- Team capability: Consider whether developers can review, test and maintain the code, including code produced with an assistant.
- Project conventions: Check how the language’s types, tests and tooling fit the interfaces and quality controls already in place.
- Model familiarity: Treat fluency as a potential convenience, not a guarantee of correctness or a measured universal advantage.
- Evidence: When evaluating a trend, distinguish platform contribution rankings, code-push rankings, repository activity and survey-reported adoption.
The available evidence shows AI use alongside changes in language activity and growing Python adoption in particular measures. It does not establish how much AI has caused language use to change across the software industry, or identify one universally best language for AI-assisted development.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
