AI Code Quality: Hype vs Reality
Itamar Friedman from Qodo reveals the hidden quality crisis in AI-generated code: why 67% of developers are concerned, testing doubles trust, and context matters more than models.
•The State of AI Code Quality
Developers Using AI Tools
82%
Using AI dev tools daily or weekly
Code AI-Generated
60%
Of developers say 25%+ of code is AI-generated
Productivity Boost
3x
In writing code with AI assistance
Quality Concerns
67%
Have serious concerns about AI code quality
Executive Summary: The Hidden Quality Crisis
While AI tools have achieved mass adoption—with 82% of developers using AI tools daily or weekly and 60% reporting 25%+ of code is AI-generated—a troubling quality crisis has emerged. Despite 3x productivity gains in writing code, 67% of developers have serious quality concerns about AI-generated code.
The data reveals a paradox: faster code creation equals more bugs. Teams spend 90% more time reviewing PRs while facing 97% more PRs and 3x more security incidents. The solution isn't fewer AI tools—it's smarter ones. Teams that heavily use AI for testing double their trust in AI-generated code, and AI code review delivers 47% productivity improvements.
The key insight: Quality is your competitive edge. AI is a tool, not a solution. Invest in quality gates, testing, and context to realize the full productivity promise.
Key Insights
Critical findings from the data and what they mean for your team
The Productivity-Quality Paradox
AI delivers 3x productivity gains in writing code, but creates a quality crisis. 90% more time to review PRs, 97% more PRs being opened, and 3x more security incidents reported. The paradox: faster code creation equals more bugs to catch.
Notable Quotes:
"We saw 3x productivity boost in writing code. But that doesn't mean that if you have 3x productivity in writing code that you actually guarantee any quality."
"Eventually it takes more time to review PR like 90% more time to review PR and by the way like there's not less amount of bugs per line of code but even if there's not less bugs per line of code you have much more bugs because there are much more PRs much more code being generated."
Testing Is the Trust Multiplier
Developers who heavily use AI for testing double their trust in AI-generated code. Testing shifts from after-the-fact validation to proactive quality assurance, becoming essential for AI workflows.
Notable Quotes:
"When they heavily use AI to do testing they actually double their trust in the AI generated code."
"The next suspect to help us with the quality is code review. What really interesting about code review is that it's a process that helps almost with all the process level and the code level like issues."
Context Is the Missing Link
80% of developers don't trust the context LLMs have. When Qodo connected their context engine, it became the #1 tool - 60% of all code generator/review calls were to context MCPs. Context matters more than the model itself.
Notable Quotes:
"Context is extremely important. As Qodo, one of our technology moats is around context and when you connect our context engine we're seeing it as the number one tool that is being used like 60% of code generator or code review tools 60% of their calls to an MCP would be to a context MCP."
"80% of developers don't trust the context LLMs have."
Code Review Automation Benefits
AI code review delivers 47% improvement in productivity of writing code. Unlike code generation which creates volume, code review focuses on quality - catching issues that humans miss and ensuring standards compliance.
Notable Quotes:
"47% improvement in productivity of writing code with AI code review."
"What really interesting about code review is that it's a process that helps almost with all the process level and the code level like issues."
The Quality Crisis: 4 Critical Challenges
Why AI-generated code is creating more problems than it solves
Security Vulnerabilities
3x more security incidents reported with AI-generated code. From "vibe coding" to "vibe reviewing" - skipping denial of service checks in security reviews.
Review Bottlenecks
90% more time to review PRs with AI-generated code. Volume of PRs increased 97%, creating review backlogs and quality gate fatigue.
High Severity Issues
17% of PRs include high severity issues. Without proper quality gates, AI-generated code introduces critical bugs faster than teams can fix them.
Context Gaps
80% of developers don't trust the context LLMs have. Poor context leads to hallucinations and irrelevant code suggestions.
4 Proven Solutions
How leading teams are closing the quality gap
Invest in AI-Powered Testing
Teams that heavily use AI for testing double their trust in AI-generated code. Shift testing left and make AI part of your quality workflow.
Automate Code Review
AI code review delivers 47% productivity improvement while catching issues that humans miss. Use AI to enforce standards and catch bugs before merge.
Prioritize Context Quality
Context is more important than model choice. 60% of tool calls are to context MCPs. Invest in rich, relevant context for better AI outputs.
Quality as Competitive Edge
Quality is your competitive edge. AI is a tool, not a solution. Invest in quality gates and testing to realize the full 2x productivity promise.
Notable Quotes from the Talk
Now people are using AI to do vibe coding but actually they're even doing it for vibe checking vibe reviewing. This is the command of Claude code for security review. It was hyped like two months ago. You are a senior security engineer. And then somewhere there down the line it says please exclude denial of service. Don't catch denial of service issues. Maybe that's part of the reason like we're having cloud outages.
60% of developers say that like quarter of their code is either generated by AI or in like shaped by AI and 15% say that even more than 80% of their code is basically generated or shaped by AI.
We started with code generation. We like out of the box use it autocomplete etc. and you invest in it and you can get more out of it. But there's a glass ceiling for how much productivity you can get from code generation.
We saw 3x productivity boost in writing code. But that doesn't mean that if you have 3x productivity in writing code that you actually guarantee any quality.
The path forward is quality is your competitive edge over your competition. AI is a tool. It's not a solution. And don't only think about code generation as the only thing. Look on the entire SDLC or product development life cycle.
Quality. You would need to invest in that. It's not out of the box. And then you would see eventually the promised 2x that that probably promised to the CEO or something like that once they give you the budget for the relevant tools.
Context is extremely important. As Qodo, one of our technology moats is around context.
When they heavily use AI to do testing they actually double their trust in the AI generated code.
Actionable Takeaways
What different audiences should do right now
For Engineering Leaders
- Invest in quality gates and testing infrastructure - it's not out of the box
- Implement AI code review to catch issues before merge (47% productivity boost)
- Prioritize context quality over model selection for better AI outputs
- Measure both code generation volume AND quality metrics
- Treat quality as competitive advantage, not cost center
For Individual Developers
- Don't rely solely on AI for security reviews or critical code paths
- Use AI-powered testing to double your trust in generated code
- Invest time in providing rich context to AI tools
- Always review AI-generated code with skepticism (90% more review time)
- Learn to work alongside AI, not blindly trust it
For AI Tool Builders
- Context is king - 60% of calls are to context MCPs
- Build quality-first features, not just code generation
- Focus on the entire SDLC, not just writing code
- Make testing and review integral to your platform
- Help users provide better context to your models
About the Speaker
Itamar Friedman
CEO & Co-founder of Qodo (formerly Codota)
Itamar Friedman is the CEO and Co-founder of Qodo, a company focused on AI-powered code quality and testing solutions. With deep expertise in developer tools and AI engineering, he leads a team that's addressing the critical gap between AI code generation and production-ready code quality.
AI code quality & testing platformThe Key Message: Quality Is Your Competitive Edge
"The path forward is quality is your competitive edge over your competition. AI is a tool. It's not a solution."
Itamar Friedman's message is clear: AI delivers massive productivity gains, but without investment in quality gates, testing, and context, you're just shipping bugs faster. The teams that succeed are those that invest in quality infrastructure alongside AI adoption.
The Formula for Success:
Invest in quality → Realize 2x productivity → Beat competition