How to Ensure Developers Use AI Assistants Securely - And Why Post-Training Surveys Are the Key

AI coding assistants like Copilot, Claude, and ChatGPT are transforming software development. According to recent surveys, over 90% of developers are already using these tools to write code, debug, and even learn new skills. But as AI becomes embedded in the developer workflow, a critical question emerges for AI leaders: how do we ensure developers are using these tools securely?

The answer lies not just in training - but in measuring the impact of that training. Post-training developer surveys are a powerful, often overlooked tool that can help organizations validate secure coding knowledge, identify gaps, and ensure that AI is accelerating development without introducing risk.

Why Post-Training Surveys Matter in the Age of AI

Security Journey, a leader in secure code training, emphasizes that training alone isn’t enough. Without feedback, organizations risk delivering content that’s too basic, too advanced, or simply ineffective. A well-designed post-training survey helps you:

• Measure awareness and knowledge growth
• Assess how relevant and applicable the training was to developers’ daily work
• Understand how often developers apply secure coding principles
• Evaluate how confident developers feel using AI tools securely
• Identify areas for improvement in both training and AI usage policies

This feedback loop is essential for continuous improvement - and for ensuring that developers are not just checking boxes, but actually changing behavior.

The Security Risks of AI Coding Assistants

While AI tools offer speed and convenience, they also introduce new risks:

• Data leakage: Code sent to external servers may expose sensitive information
• Model training risks: Proprietary code could be used to train public models
• Insecure suggestions: AI-generated code may include vulnerabilities
• Overreliance: Developers may trust AI output without proper validation

In fact, nearly half of developers surveyed by CodeSignal expressed concerns about the security and privacy of AI-generated code. That’s why secure coding training must now include modules on AI-specific risks - like prompt injection, data leakage, and insecure plugin design.

Embracing AI Securely, Without Slowing Down

At our core, we believe in empowering developers to embrace AI - not avoid it. When used responsibly, AI coding assistants can dramatically improve productivity, reduce repetitive work, and even help developers learn secure coding practices. But this only works when developers are trained to use these tools safely and understand their limitations.

Surveys help you assess whether developers are confident using AI tools, whether they’re applying secure practices, and whether your training is keeping pace with the evolving AI landscape. They also help you uncover friction points - like unclear policies or lack of support - that might be holding developers back from fully leveraging AI.

What to Ask in Your Developer Survey

To help you get started, we’re offering our Secure Code Training – Developer Survey as a free resource. It includes questions across key categories:

You’ll also find open-ended questions to capture qualitative feedback - like what developers found most valuable, or what they’d like to see improved.

Download the Survey and Start Measuring What Matters

If you’re serious about enabling secure, high-velocity development with AI, you need to know what’s working - and what’s not. Post-training surveys give you the insights to:

• Benchmark progress over time
• Tailor future training to developer needs
• Strengthen your AI usage policies

• Build a culture of secure, empowered development

Download the Secure Code Training – Developer Survey now and use it with your own teams. Whether you’re using Security Journey or another training platform, this survey can be easily adapted to fit your needs.

By listening to your developers and acting on theirfeedback, you’ll ensure that your organization can move fast - and stay secure.