跳到内容

Start Date

April 8, 2026

End Date

May 8, 2026

Respondents

7,258

Survey Archive

Review Questions →

Get Data

Download →

Whether you're pro- or anti-AI, I think most of us agree that AI is having a huge impact on our work as developers, and beyond that, on the world at large.

The 2026 edition of the State of Web Dev AI survey collected data from 7,258 developers, providing a lot of fascinating material to study this impact. And while I would certainly recommend checking out the full survey results at your own pace, here are four key trends to get started exploring the data.

#1: Growing AI Adoption

It's important to acknowledge that open surveys such as this one can suffer from inherent selection bias. In other words, an AI-focused survey is probably more likely to be filled out by developers who already use and care about AI.

With that caveat in mind, if we assume that the trends revealed here apply to the industry at large then AI-assisted coding's transition from an early-adopter experiment to a standard practice is now well underway.

This becomes clear when looking at the proportion of AI-generated code produced by respondents, which has jumped from 28% on average in 2025 to 54% this year, with the 75%+ segments seeing the highest growth (delta from the previous year shown by striped bar overlays).

AI 代码生成

您编写的代码中有多少比例是 AI 生成的?
0%
20%
40%
60%
80%
100%
1
0% AI
2
|
3
|
4
|
5
50%
6
|
7
|
8
|
9
100% AI
0%
20%
40%
60%
80%
100%
Similarly, respondents are also using AI much more often, with the percentage of respondents stating they use AI “constantly” doubling year-over-year.

代码生成频率

您多久使用一次 AI 工具来生成或重构代码?
0%
20%
40%
60%
80%
100%
1
从不
2
每年几次
3
每月几次
4
每周几次
5
每天几次
6
每小时几次
7
经常
0%
20%
40%
60%
80%
100%

#2: Claude Code & the Rise of Coding Agents

This adoption is in large part driven by a new generation of coding agents and assistants, with Claude Code taking the lead when looking at positive respondent sentiment regarding coding agents.

通过...分组:

根据此项排序:

1
[zh-Hans] tools.coding_agents_assistants
42.3%
47.9%
9.3%
2
[zh-Hans] tools.coding_agents_assistants
26.3%
59.8%
13.4%
3
[zh-Hans] tools.coding_agents_assistants
22.6%
54.9%
21.7%

经验

  • <span aria-hidden="true">🤓</span> 用过: 使用过此项目的受访者。
  • <span aria-hidden="true">👀</span> 有听说过: 听说过此项目,但没有使用过此项目的受访者。
  • <span aria-hidden="true">🤷</span> 没听说过: 从未听说过此项目的受访者。

情绪(态度)

  • 积极: 有兴趣了解此技术,或者愿意再次使用它的受访者
  • 中立: 未对此技术发表看法的受访者
  • 消极: 对此技术不感兴趣,或者使用此技术有糟糕体验的受访者

These agents are emerging as a more convenient way to interact with LLMs, and potentially replacing chatbots, app generators, and other specialized tools.

As a consequence, despite ChatGPT having the edge in terms of sheer popularity, Claude is the model that developers actually pay for the most.

Paid Model Usage
✨

Out of the models mentioned here, which one do you or your company actually pay for?
多选
0%
20%
40%
60%
80%
100%
04
无
0%
20%
40%
60%
80%
100%

#3: Race to Monetization

Speaking of paying more: now that we've all gotten hooked on cheap AI subsidized by unprecedented VC funding rounds, our collective addiction is making it easier for AI labs to start raising prices. And this translates into a clear increase in individual AI spend compared to last year.

您个人每月在 AI 工具上花费多少(美元)?
0%
20%
40%
60%
80%
100%
1
$0
2
$1-$20
3
$20-$50
4
$50-$100
5
$100-$500
6
$500-$1000
7
$1000-$5000
8
过滤掉的答案
-2
0%
20%
40%
60%
80%
100%
Only time will tell if this increased monetization will justify the current eye-popping AI company valuations, or if they instead mean we're in the midst of an AI bubble. But either way, survey respondents certainly think that we are.
We're currently in an “AI bubble”
0%
20%
40%
60%
80%
100%
1
强烈反对
2
反对
3
中立
4
同意
5
十分同意
0%
20%
40%
60%
80%
100%

#4: Increased Risk Factors

It would be irresponsible to talk about AI progress without also mentioning the many issues stemming from all these advances.

There is a real risk that many of us will lose our livelihood because of AI improvements. After all, whether AI can actually do our jobs or not, all that's required for developers to be replaced is that management believes it can.

AI 工具对我的工作保障构成威胁
0%
20%
40%
60%
80%
100%
1
强烈反对
2
反对
3
中立
4
同意
5
十分同意
0%
20%
40%
60%
80%
100%

But the list of AI risks does not stop there.

Right now, armed forces throughout the world are working on autonomous drones and targeting systems that could lead to diminished responsibility and accountability; and this is of course happening against a backdrop of ever-increasing AI resource consumption and accelerating climate change.

Which of these general AI risks and issues are you most concerned about?
多选
Limited
3
0%
20%
40%
60%
80%
100%
01
Job displacement
02
Military use of AI
03
Environmental impact
04
AI slop takeover
05
Negative cognitive impacts
06
Security issues
07
AI overuse
08
Rising AI costs
09
Copyright and ethics issues
10
Open-source disruption
0%
20%
40%
60%
80%
100%
Even if you discount these externalities, as we rely on AI more and more, it's hard to shake the feeling that maybe our new cyber-buddies aren't being fully truthful with us, with hallucinations and inaccuracies being cited as the most common AI pain points.
在使用 AI 工具时,您最大的痛点是什么?
多选
Limited
3
0%
20%
40%
60%
80%
100%
01
幻觉和不准确
+8
02
代码质量
-1
03
缺乏上下文
-1
04
财务成本
+3
05
隐私担忧
-1
06
提示词理解
-3
07
版权担忧
-2
08
集成困难
+1
09
响应时间慢
10
道德伦理担忧
11
缺乏对提示词的理解
12
生成的代码质量差
13
其他答案
0%
20%
40%
60%
80%
100%

The AI Future Is Here Whether We Want It or Not

I'll be honest: I'm not sure where all this leaves us. But whether you think we should be embracing AI, fighting it, or maybe something in between, one thing's for sure: it's becoming increasingly difficult to ignore it.

So if like me you haven't quite made up your mind on this complex topic, hopefully these survey results can be one more data point that will help you make more informed decisions–without having to ask ChatGPT first.

我们的合作伙伴

在日本找到你的梦想开发者工作。
通过深入的现代前端工程课程提升您的技能。
直接从早期采用者和技术作者那里了解最新趋势。
The application monitoring platform loved by developers worldwide.
Full stack auth and user management that lets you launch faster and scale easier.