90% of professional developers now regularly use at least one AI tool at work, and 74% have adopted specialized AI coding tools — not just general chatbots. Those are the headline numbers from JetBrains’ January 2026 AI Pulse survey of over 10,000 developers across eight languages and multiple continents, the most credible real-work adoption data available today.
The JetBrains AI Pulse Survey: Why This Data Matters
The JetBrains AI Pulse survey, conducted in January 2026 with over 10,000 professional developers across 8 languages and globally representative sampling, is the benchmark dataset for understanding AI coding tool adoption. Unlike vendor-reported user counts or opt-in web surveys, JetBrains used raking weighting to ensure the sample matched the global developer population — making it the most methodologically rigorous independent survey on this topic. JetBrains tracked the same metrics across multiple survey waves (April 2025, June 2025, January 2026), enabling rare longitudinal trend analysis. The survey separated “awareness” from “work adoption,” a distinction that eliminates the noise of casual experimentation and surfaces tools developers actually trust enough to use professionally. This data reveals which tools have earned real slots in developer workflows versus which are popular in demos but abandoned in production. For any developer or engineering leader trying to make a budget or tooling decision in 2026, the JetBrains AI Pulse is the most reliable starting point — not vendor marketing, not Twitter discourse, and not smaller single-country surveys.
What Makes the JetBrains Methodology Different?
The JetBrains AI Pulse separates “awareness” from “work adoption” — a critical distinction. A tool can be famous (high awareness) while almost nobody uses it professionally (low adoption). The survey also tracked longitudinal change across three waves (April 2025, June 2025, January 2026), giving it rare time-series data on growth trajectories. The sample of 10,000+ developers in 8 languages makes it globally representative in a way that English-only surveys cannot achieve. JetBrains’ existing Developer Ecosystem Survey (24,534 developers across 194 countries) further validates the context.
The Big Picture — 90% of Developers Now Use AI at Work
AI coding tool adoption has crossed a threshold in 2026: 90% of professional developers regularly use at least one AI tool at work, and 74% have adopted specialized tools built specifically for developers — not just general-purpose chatbots. This is a seismic shift from 2024, when Stack Overflow data showed 76% using or planning to use AI tools. The JetBrains data makes the adoption curve concrete: in less than two years, AI tools went from an experimental novelty to a professional baseline expectation. The Sonar State of Code Survey 2026 (1,100 developers) adds that 72% of those who have tried AI now use it nearly every day. Stack Overflow’s 2025 survey of 49,000+ developers found 47.1% use AI tools daily, and only 16.2% have no plans to use AI tools at all. The 10-percentage-point drop in non-adopters between 2024 and 2025 alone signals that “I don’t use AI tools” is becoming professionally unusual rather than the default. For engineering leaders, this means AI coding tools are now infrastructure — not innovation projects.
Who Is Still Not Using AI Tools?
The 10% who don’t use AI tools cluster in regulated industries with strict data governance requirements — defense, healthcare with specific compliance mandates, and some financial services firms where IP leakage risk outweighs productivity benefits. Notably, even in these environments the trajectory is toward adoption as on-premise and air-gapped options mature. JetBrains data does not yet break non-adopters down by industry, but enterprise patterns (discussed later) point to security and compliance as the primary blockers, not preference.
Tool-by-Tool Breakdown: Awareness vs. Actual Work Adoption
The JetBrains AI Pulse data reveals a stark awareness-to-adoption gap that separates genuinely useful tools from marketing-inflated ones. As of January 2026: GitHub Copilot sits at 76% developer awareness but only 29% work adoption — a 47-point gap. Cursor has 69% awareness and 18% work adoption. Claude Code shows 57% awareness and 18% work adoption, but its awareness is still growing fast, meaning its conversion rate is accelerating. In contrast, general-purpose chatbots used for coding show different patterns: ChatGPT is used for coding by 28% of developers, making it the most common AI coding tool by raw usage. Google Antigravity, launched in November 2025, already hit 6% adoption in its first two months. JetBrains AI Assistant holds 9% and Junie 5%. The conversion analysis from independent research shows Cursor has the highest conversion rate among aware developers at 35.5% — meaning that among developers who know about Cursor, more than one-third are using it professionally.
| Tool | Awareness | Work Adoption | Gap |
|---|---|---|---|
| GitHub Copilot | 76% | 29% | 47 pts |
| Cursor | 69% | 18% | 51 pts |
| Claude Code | 57% | 18% | 39 pts |
| ChatGPT (coding) | N/A | 28% | — |
| JetBrains AI Asst. | N/A | 9% | — |
| Google Antigravity | N/A | 6% | — |
| Junie | N/A | 5% | — |
Reading the Awareness-Adoption Gap
A narrow awareness-adoption gap signals high conversion efficiency — developers who hear about a tool are likely to adopt it professionally. Claude Code’s 39-point gap is the narrowest of the three leading specialized tools, suggesting that word-of-mouth discovery (which is how most Claude Code awareness spreads) leads more reliably to professional use than paid marketing (Copilot’s playbook). GitHub Copilot’s 47-point gap reflects the reality that many developers know about it from GitHub integration but don’t find it compelling enough to use regularly at work.
Claude Code’s Explosive 6x Growth — What’s Driving It?
Claude Code grew from approximately 3% work adoption in April-June 2025 to 18% globally in January 2026 — a 6x increase in about six months. In the United States and Canada, Claude Code reached 24% adoption, making it the second most-adopted specialized AI coding tool in North America. This is likely the fastest adoption trajectory in the history of professional developer tooling. The growth is not driven by marketing spend: Claude Code gained visibility primarily through developer word-of-mouth, particularly on GitHub discussions, Hacker News, and developer-focused YouTube. The product’s quality metrics explain the retention: Claude Code holds the highest customer satisfaction of any specialized AI coding tool — CSAT of 91% and Net Promoter Score of 54. For context, NPS above 50 is considered exceptional in B2B software. No other specialized AI coding tool has published equivalent satisfaction figures, and the JetBrains survey independently confirms these loyalty metrics through its own retention analysis. Claude Code’s growth demonstrates the “performance over platform” hypothesis: when a tool is materially better at the core task, developers will abandon ecosystem lock-in (VS Code extensions, GitHub integration) to use it.
Why Developers Switch to Claude Code
Based on the survey data and public developer feedback, the primary drivers of Claude Code adoption are: (1) autonomous multi-file editing capability that outperforms competitors on complex refactoring tasks; (2) deeper Claude 3/4 model integration compared to competitors using OpenAI models; (3) CLI-first workflow that fits senior developers who live in the terminal; and (4) lower hallucination rates on large-codebase tasks, which reduces the debugging overhead that makes 70% of developers lose time to AI code review. The 6x growth rate means Claude Code is not just attracting new AI tool users — it is pulling developers who already tried Copilot or Cursor and found them insufficient.
GitHub Copilot: Still the Market Leader, But Growth Has Stalled
GitHub Copilot remains the most widely adopted specialized AI coding tool at 29% work adoption globally, but JetBrains’ longitudinal data shows its growth trajectory has flattened significantly compared to prior survey waves. The 76% awareness figure means nearly all professional developers know about Copilot, yet fewer than two in five aware developers are using it professionally — suggesting that the remaining non-adopters have already evaluated and passed on the tool. Where Copilot maintains a structural advantage is enterprise: in companies with over 5,000 employees, Copilot reaches 40% adoption, driven by enterprise agreements with GitHub (now part of Microsoft) and procurement cycles that favor established vendors. For individual developers and smaller teams, Copilot’s lead has effectively disappeared — Claude Code and Cursor have closed the gap to statistical equivalence at 18% each. The Sonar State of Code Survey 2026 (1,100 developers) shows slightly different tool share numbers (Copilot 75%, ChatGPT 74%, Claude 48%, Cursor 31%) but used different methodology and definitions, suggesting real-world tool usage is blending — developers use multiple tools, not one exclusively.
What Copilot Still Does Well
GitHub Copilot’s integration with GitHub and VS Code remains frictionless for developers already on those platforms. Enterprise procurement, compliance certifications, and SOC 2 documentation give Copilot an institutional footprint that Claude Code has not yet matched. For teams where AI tool selection is a top-down IT decision rather than a bottom-up developer choice, Copilot’s enterprise credibility matters more than satisfaction scores. Its 29% adoption is a plateau, but it is a high plateau — representing millions of professional developers globally.
The Trust Gap: High Adoption, Low Confidence in AI Output
The most striking finding across all 2026 developer surveys is the simultaneous rise of AI tool adoption and erosion of confidence in AI-generated code. Sonar’s 2026 State of Code Survey found that 96% of developers do not fully trust that AI-generated code is functionally correct. Stack Overflow and Harness Research found only 32.7% of developers trust AI output, while 45.7% actively distrust it. Modall AI’s 2026 report found AI-generated code contains 2.74 times more vulnerabilities than human-written code. Despite this, 48% of developers don’t always review AI code before committing it. Stack Overflow data shows 76% of developers say AI increases productivity — but 70% also report spending extra time debugging AI-generated code. This paradox defines the 2026 developer experience: tools that accelerate first-draft velocity while creating a new category of review and debugging work. The implication for engineering teams is that AI coding tool ROI is not automatic — it requires review practices, CI code quality gates, and prompting discipline to capture the productivity gain without importing technical debt.
What Developers Distrust Most
JetBrains data and Sonar’s survey converge on three categories of AI code quality concerns: unreliability (53% of developers who reported issues), duplicated logic (40%), and security vulnerabilities (57% cite sensitive data exposure concerns). The 2.74x vulnerability rate in AI-generated code — measured by static analysis across large codebases — suggests that teams using AI tools without static analysis review are accepting systematically higher security risk. The practical mitigation is not to stop using AI tools but to integrate them with code quality scanning at the PR level rather than relying on developer review alone.
Enterprise vs. Individual: How Adoption Splits by Company Size
The JetBrains survey reveals a clear bifurcation in AI coding tool adoption by company size. GitHub Copilot reaches 40% adoption at companies with 5,000+ employees — more than double its global average — driven by enterprise software agreements and centralized procurement. Smaller companies and individual developers (freelancers, startup founders, side-project developers) show the reverse pattern: Claude Code and Cursor adoption in this segment substantially exceeds their global averages, based on the North American adoption data and public developer community signals. The JetBrains Developer Ecosystem Survey 2025 adds context: 68% of developers expect employers to require proficiency in AI tools in the near future, and nearly 9 in 10 already save at least one hour per week using AI tools — with 1 in 5 saving 8 or more hours per week. For enterprises making tooling decisions, the data suggests a bifurcated strategy: Copilot for compliance-sensitive, GitHub-integrated environments; Claude Code or Cursor for engineering teams optimizing for developer satisfaction and retention.
The Skill Premium Emerging Around AI Tools
The 68% of developers who expect AI tool proficiency to become a job requirement points to an emerging skills gap. Developers who master effective prompting, multi-agent workflow design, and AI code review are already commanding salary premiums — a trend visible in job listings and compensation surveys from early 2026. Gartner’s 2026 forecast adds a macro lens: 40% of enterprise applications will incorporate AI agents by the end of 2026, creating demand for developers who understand how to build with and oversee AI systems, not just use them as autocomplete.
How the AI Coding Tool Market Will Shift in 2026
Three forces will reshape AI coding tool market share through the rest of 2026. First, Google Antigravity’s fast start — 6% adoption in two months after launching in November 2025 — shows that a well-resourced new entrant with deep IDE integration can acquire developer share quickly. Google’s ability to embed Antigravity into Android Studio, the default Android development environment, and link it to Gemini’s model improvements gives it a structural distribution advantage that neither Cursor nor Claude Code can easily match. Second, the agent capability gap is widening: Gartner projects 40% of enterprise apps will incorporate AI agents by end of 2026, and the tools that best support autonomous multi-step agentic workflows (where Claude Code currently leads) will capture disproportionate enterprise spend. Third, the trust gap creates an opening for tools that combine generation with integrated code quality scanning — a combination no major tool has fully solved as of mid-2026. The Sonar data (96% don’t trust AI code, 48% don’t review before committing) points to a product gap that AI-native static analysis tools are beginning to address.
Will the Market Consolidate?
The current trajectory suggests consolidation to three primary tools — GitHub Copilot, Claude Code, and Cursor — with a long tail of specialized tools for specific IDEs or languages. JetBrains AI Assistant’s 9% and Junie’s 5% suggest JetBrains is building adoption within its own IDE ecosystem (IntelliJ, PyCharm, WebStorm), which could stabilize its position without threatening the top three. Google Antigravity’s trajectory over the next two quarters will determine whether there are four leaders or three.
Choosing the Right AI Coding Tool Based on Survey Data
The JetBrains AI Pulse data and corroborating surveys provide a framework for choosing AI coding tools based on actual adoption and satisfaction data rather than marketing. For individual developers and small teams prioritizing satisfaction and code quality: Claude Code’s 91% CSAT and 54 NPS are the strongest signals available — no other tool has demonstrated this loyalty in independent survey data. Its 6x growth suggests a tool with genuine product-market fit, not just marketing spend. For teams deeply integrated with GitHub and VS Code: GitHub Copilot’s 29% adoption and enterprise compliance certification make it the lowest-friction option, particularly in organizations where procurement runs through existing Microsoft agreements. For developers who want IDE-replacement depth: Cursor’s 35.5% conversion rate among aware developers — the highest in the market — indicates that developers who try Cursor tend to stay. Its AI-native editor design (rather than plugin approach) appeals to developers who want AI integrated at the editor level, not bolted on.
| Factor | Best Choice |
|---|---|
| Highest satisfaction (CSAT/NPS) | Claude Code |
| Enterprise procurement / GitHub integration | GitHub Copilot |
| IDE-native AI experience | Cursor |
| Android / Google Cloud stack | Google Antigravity |
| JetBrains IDE users | JetBrains AI Assistant |
The data also suggests a practical starting point: if you haven’t yet adopted a specialized AI coding tool, the survey data recommends starting with Claude Code or Cursor based on satisfaction scores and growth trajectory, then evaluating whether GitHub Copilot’s GitHub integration offers enough workflow benefit to switch. Don’t start with Copilot just because it’s the most-known tool — awareness doesn’t equal satisfaction.
FAQ
The following questions are answered using data from the JetBrains AI Pulse survey (January 2026, 10,000+ developers), Stack Overflow Developer Survey 2025 (49,000+ respondents), Sonar State of Code Survey 2026 (1,100 developers), and JetBrains Developer Ecosystem Survey 2025 (24,534 developers across 194 countries). These four surveys represent the most rigorous independent data available on AI coding tool adoption, satisfaction, and developer behavior as of mid-2026. Together they paint a consistent picture: adoption is nearly universal among professional developers, satisfaction is high for a narrow set of tools, trust in AI-generated code remains limited despite daily use, and the market is consolidating around three to four primary tools with distinct enterprise versus individual adoption patterns. Key statistics are cited to their source for accuracy. If a figure differs between surveys, both numbers are provided because each survey used different sampling, definitions, and question framing.
What percentage of developers use AI coding tools in 2026?
According to the JetBrains AI Pulse survey of 10,000+ developers (January 2026), 90% of professional developers regularly use at least one AI tool at work, and 74% have adopted specialized AI coding tools (coding assistants, AI editors, or agents). Stack Overflow’s 2025 survey (49,000+ respondents) found 84% use or plan to use AI tools, up from 76% in 2024.
Which AI coding tool has the highest adoption in 2026?
GitHub Copilot leads specialized AI coding tool adoption at 29% globally, with a spike to 40% in companies with 5,000+ employees. However, ChatGPT used for coding tasks reaches 28% — close to Copilot’s figure — because developers use it informally even when they have a specialized tool. Claude Code and Cursor are tied at 18% global work adoption.
What is Claude Code’s growth rate in 2026?
Claude Code grew approximately 6x from ~3% work adoption in April-June 2025 to 18% globally by January 2026 — the fastest adoption trajectory recorded in professional developer tooling history. In the US and Canada, Claude Code reached 24% adoption. It holds the highest customer satisfaction scores: CSAT 91% and NPS 54.
Do developers trust AI-generated code?
No — 96% of developers do not fully trust AI-generated code, according to Sonar’s 2026 survey of 1,100 developers. Only 32.7% of developers trust AI output, while 45.7% actively distrust it (Stack Overflow / Harness). Despite this, 48% don’t always review AI code before committing. AI-generated code has been found to contain 2.74x more vulnerabilities than human-written code.
How much time do developers save with AI coding tools?
JetBrains’ Developer Ecosystem Survey 2025 (24,534 developers) found nearly 9 in 10 developers save at least one hour per week using AI tools, and 1 in 5 save 8 or more hours per week. However, 70% of developers also report spending extra time debugging AI-generated code — so the net time savings depends heavily on review practices and tool quality.