[Survey of 108 AI-Utilizing Engineers] Is 'AI Stealing Jobs' True? — 90% Positively Embrace Workflow Changes; 'Ability to Evaluate AI Output' Outweighs 'Coding Ability'
TWOSTONE&Sons surveyed 108 developers who use AI tools. Nearly 90% view AI-driven workflow changes positively, citing significant time savings in testing and coding. However, new responsibilities such as prompt engineering and AI output review are emerging, shifting requirements toward mastering AI orchestration.
📋 Article Processing Timeline
- 📰 Published: May 21, 2026 at 20:00
- 🔍 Collected: May 21, 2026 at 11:31
- 🤖 AI Analyzed: May 22, 2026 at 05:36 (18h 4m after Collected)
## Survey Background and Purpose
TWOSTONE&Sons conducted a survey on 108 software developers who have used AI tools (such as GitHub Copilot, Claude, and ChatGPT) or AI agents in their work within the past six months to understand how AI is transforming development workflows.
## Key Findings
### 1. Reduction and Efficiency in Workflow
Respondents reported significant time savings in tasks such as 'Test execution/result confirmation' (38.9%), followed by 'Test code creation' (38.0%) and 'Standardized coding, such as CRUD operations' (38.0%). This demonstrates that AI is successfully alleviating the manual burden of routine tasks.
### 2. Emerging New Responsibilities
Conversely, 72.2% of engineers noted that new or increased responsibilities have emerged. These include 'Prompt design/optimization' (52.6%), 'Integration and refactoring of AI-generated code' (41.0%), and 'Review and quality assurance of AI output' (38.5%), indicating that the use of AI introduces new management and coordination requirements.
### 3. Shift in Awareness and Positive Reception
Approximately 90% of engineers view these changes positively. Engineers feel that the 'breadth of required skills has expanded' (53.8%), signaling that they welcome the change and perceive it as an opportunity to extend their capabilities.
### 4. Impact on Project Management
Changes in project delivery were also reported, including 'Increase in short-term projects' (38.0%), 'Improved development speed' (36.1%), and 'Cases of simultaneous multi-project handling' (34.3%). AI adoption is not only enhancing individual productivity but also driving agility across entire projects.
## Survey Methodology
- Title: Survey on Workflow Transformation for AI-Utilizing Development Engineers
- Method: Internet survey planned by 'Re-a-py', a research marketing platform provided by IDEATECH
- Period: May 7 – May 8, 2026
- Participants: 108 developers who have utilized AI in their work within the last six months
Source: TWOSTONE&Sons Co., Ltd. (Stock Code: 7352)
TWOSTONE&Sons conducted a survey on 108 software developers who have used AI tools (such as GitHub Copilot, Claude, and ChatGPT) or AI agents in their work within the past six months to understand how AI is transforming development workflows.
## Key Findings
### 1. Reduction and Efficiency in Workflow
Respondents reported significant time savings in tasks such as 'Test execution/result confirmation' (38.9%), followed by 'Test code creation' (38.0%) and 'Standardized coding, such as CRUD operations' (38.0%). This demonstrates that AI is successfully alleviating the manual burden of routine tasks.
### 2. Emerging New Responsibilities
Conversely, 72.2% of engineers noted that new or increased responsibilities have emerged. These include 'Prompt design/optimization' (52.6%), 'Integration and refactoring of AI-generated code' (41.0%), and 'Review and quality assurance of AI output' (38.5%), indicating that the use of AI introduces new management and coordination requirements.
### 3. Shift in Awareness and Positive Reception
Approximately 90% of engineers view these changes positively. Engineers feel that the 'breadth of required skills has expanded' (53.8%), signaling that they welcome the change and perceive it as an opportunity to extend their capabilities.
### 4. Impact on Project Management
Changes in project delivery were also reported, including 'Increase in short-term projects' (38.0%), 'Improved development speed' (36.1%), and 'Cases of simultaneous multi-project handling' (34.3%). AI adoption is not only enhancing individual productivity but also driving agility across entire projects.
## Survey Methodology
- Title: Survey on Workflow Transformation for AI-Utilizing Development Engineers
- Method: Internet survey planned by 'Re-a-py', a research marketing platform provided by IDEATECH
- Period: May 7 – May 8, 2026
- Participants: 108 developers who have utilized AI in their work within the last six months
Source: TWOSTONE&Sons Co., Ltd. (Stock Code: 7352)
FAQ
AI導入で最も業務時間が減ったタスクは何ですか？
調査の結果、「テスト実行・結果確認」が38.9%で最多となりました。次いで「テストコードの作成」および「定型的なコーディング」が各38.0%となっています。
AI導入後に新たに発生・増加した業務は何ですか？
最も多いのは「プロンプトの設計・最適化」で52.6%でした。次いで「AI生成コードの統合・リファクタリング」が41.0%、「AI出力のレビュー・品質担保」が38.5%と続きます。
エンジニアはAI導入による業務変化をどう捉えていますか？
約9割（87.1%）のエンジニアが業務変化をポジティブに捉えています。スキルの幅が広がったと回答するエンジニアも半数を超えています。
プロジェクトの進め方にどのような変化がありましたか？
「短期プロジェクトの増加」（38.0%）や「プロジェクトの開発スピード向上」（36.1%）、「複数プロジェクトの並行対応」（34.3%）といった変化が顕著です。
この調査の対象者は誰ですか？
GitHub CopilotやChatGPT等のAIツールやAIエージェントを、直近6ヶ月以内に開発業務で使用している開発エンジニア108名です。