Microsoft has embarked on a transformative journey, announcing a bold initiative to phase out all C and C++ code from its products by the year 2030. This ambitious plan, revealed on November 25, 2025, comes in the wake of significant operational challenges within Windows 11, which had been plagued by malfunctions since July of the same year. Galen Hunt, a Distinguished Engineer at Microsoft, articulated a clear goal: “1 engineer, 1 month, 1 million lines of code,” setting a high bar for the company’s transition to AI-assisted code rewriting.
The technical breakdown
The issues affecting Windows 11 were primarily linked to complications with XAML components, the framework utilized for constructing Windows interfaces. Users reported a range of frustrations, including unresponsive Start Menus, disappearing Taskbars, sluggish File Explorer performance, and erratic System Settings. IT professionals noted that these malfunctions resulted in significant downtime, costing enterprises valuable productivity. Gartner estimates that software bugs of this scale can lead to billions in losses annually.
In response to the widespread failures, Microsoft suggested temporary fixes, such as resetting applications or rolling back updates, which the company acknowledged were merely “bandaids.” A patch is promised for December 2025, aimed at addressing the underlying issues that have persisted for months.
As discussions unfolded on social media, many software engineers and developers expressed skepticism about Microsoft’s testing capabilities. One engineer remarked, “Microsoft’s push for feature-rich updates seems to be outpacing their testing capabilities,” highlighting a growing concern within the tech community.
Hunt’s vision for the future includes building a robust algorithmic infrastructure that can effectively manage and modify code at scale. This infrastructure is designed to enhance code understanding and facilitate the application of AI agents to streamline code modifications.
The Rust conversion plan
In line with this strategy, Hunt has opened a Principal Software Engineer position, emphasizing the need for expertise in Rust—a programming language known for its memory safety and concurrency features. The job posting specifies that candidates should possess at least three years of experience in systems-level coding, with a preference for those familiar with compiler, database, or operating system implementations.
“Our strategy is to combine AI and Algorithms to rewrite Microsoft’s largest codebases,” Hunt stated, indicating a forward-thinking approach within the Future of Scalable Software Engineering group. The team aims to tackle technical debt at scale while fostering a diverse and growth-oriented environment.
Despite the promise of Rust, some developers have voiced concerns regarding the effectiveness of AI in generating Rust code, citing limited training data compared to more established languages. “Anyone who used AI to write Rust knows LLMs are not very effective at it,” one developer commented, raising questions about the feasibility of the rewrite strategy.
Industry reaction and concerns
The announcement has sparked considerable debate within technology circles. Some industry observers have questioned the economic rationale behind the initiative, suggesting that the potential savings from AI-generated code could be overshadowed by the productivity losses stemming from system failures. “So to save a few hundred thousand dollars, they end up losing billions,” one commenter quipped, reflecting a broader skepticism about the quality of AI-generated code.
The implications of AI-generated code extend beyond Microsoft, as the industry grapples with instances where AI tools produce seemingly plausible yet fundamentally flawed implementations. Such systems often generate references to non-existent functions or libraries, presenting fabricated information with unwarranted confidence.
Microsoft’s advertising business, which has integrated AI capabilities extensively throughout 2025, provides context for the company’s overarching AI strategy. By April 2025, the company’s advertising revenue had surpassed billion annually, with significant growth in search and news advertising. The success of AI-integrated features in consumer-facing products contrasts sharply with the foundational nature of operating system components, where failures can disrupt millions of devices.
The workforce dimension
Microsoft’s AI-assisted development approach also carries significant workforce implications. The company announced plans to resume hiring with an emphasis on leveraging AI tools, suggesting anticipated productivity gains. CEO Satya Nadella likened the current moment to the introduction of transformative technologies in the 1980s and 1990s, emphasizing the need for organizations to redesign processes rather than merely automating existing workflows.
This focus on “leverage” implies that Microsoft aims to capture substantial value from AI deployment, rather than simply passing benefits to employees or consumers. The company anticipates a learning curve for employees as they adapt to new workflows, with productivity gains expected to materialize over the course of a year.
As platforms consolidate around proprietary AI systems, marketing and advertising professionals face uncertainty. Microsoft’s decision to sunset its Xandr DSP by February 28, 2026, exemplifies how AI-driven consolidation impacts specialized platforms.
Infrastructure investments
Microsoft’s commitment to AI is underscored by its unprecedented infrastructure investments, with plans to allocate billion in 2025 for AI datacenter construction globally. More than half of this investment is directed toward facilities in the United States. The completion of the Fairwater datacenter in Wisconsin, featuring hundreds of thousands of NVIDIA GB200 processors, marks a significant milestone in the company’s AI infrastructure strategy.
This facility is designed to deliver ten times the performance of the fastest supercomputers, supporting the code generation and processing capabilities that Hunt envisions. Such investments reflect Microsoft’s dedication to AI-driven development strategies, even as questions linger regarding the technology’s reliability for mission-critical systems.
The testing challenge
The announcement highlights a critical tension between the rapid pace of AI-assisted development and the rigorous testing required for operating system reliability. With millions of organizations relying on Windows for daily operations, the four-month delay in acknowledging systematic failures raises concerns about the company’s ability to detect and respond to issues effectively.
One former employee noted that the combination of a monthly update schedule and aggressive feature development may strain testing capabilities. “C/C++ requires deep understanding of memory management, pointer arithmetic, and low-level system interactions,” they observed, questioning whether AI systems can safely rewrite such complex code.
Microsoft has yet to disclose the testing methodologies that will ensure the correctness of AI-generated code at the scale Hunt described, as traditional software testing approaches often struggle with systems containing millions of lines of code.
Broader implications for the advertising industry
The developments at Microsoft resonate beyond its immediate ecosystem, as marketing professionals navigate the increasing integration of AI capabilities in advertising. The introduction of Image Animation capabilities on November 17, 2025, exemplifies the potential for AI to transform static images into dynamic video assets, enhancing performance across the Microsoft Advertising Network.
Research indicates that a significant majority of advertisers are exploring or implementing generative AI for video creation, underscoring the growing relevance of AI in marketing strategies. However, the distinction between experimental consumer features and foundational system components remains crucial, as failures in advertising platforms affect individual campaigns rather than entire business operations.
What comes next
The upcoming December 2025 patch for Windows 11 will serve as a critical test of Microsoft’s ability to resolve the issues that have persisted since July. While the patch aims to address immediate symptoms, it does not alleviate concerns about the potential for similar problems as AI-generated code becomes more prevalent.
Hunt’s job posting for Principal Software Engineers remains open, indicating that Microsoft recognizes the complexity of the transformation ahead. The requirement for in-person work in Redmond suggests that remote collaboration may not suffice for the level of coordination necessary to rewrite millions of lines of critical system code.
As Microsoft continues to navigate this ambitious initiative, its quarterly earnings reports will provide insights into the impact of AI across the business. With cloud revenue reaching .1 billion in recent quarters and AI features engaging 900 million monthly active users, the company is well-positioned to explore the potential of AI, even as it grapples with the challenges of modernizing legacy codebases.
Timeline
- April 2025: Microsoft CEO Satya Nadella announces AI writes 20-30% of company code
- July 2025: Windows 11 core components begin malfunctioning across Start Menu, Taskbar, File Explorer, and System Settings
- July 31, 2025: Microsoft search advertising revenue climbs 21% in record quarter
- August 19, 2025: Microsoft Copilot achieves 73% higher click-through rates in advertising study
- September 20, 2025: Microsoft completes Fairwater datacenter in Wisconsin with hundreds of thousands of NVIDIA GB200s
- October 2025: Windows 11 problems persist through monthly updates with worsening severity
- November 2, 2025: Microsoft announces plans to hire again with “more leverage” from AI tools
- November 17, 2025: Microsoft adds image animation and performance tracking to Copilot AI tools
- November 25, 2025: Microsoft officially acknowledges Windows 11 component malfunctions; Galen Hunt announces plan to eliminate all C and C++ code by 2030
- December 2025: Microsoft promises patch to fix Windows 11 component failures
- February 28, 2026: Microsoft Xandr DSP sunset deadline
- 2030: Target date for eliminating all C and C++ code from Microsoft products
Summary
Who: Galen Hunt, Distinguished Engineer at Microsoft, announced the initiative affecting millions of Windows users who experienced system malfunctions. CEO Satya Nadella and CTO Kevin Scott previously disclosed AI’s increasing role in code generation across Microsoft products.
What: Microsoft plans to eliminate all C and C++ code from its products by 2030 using AI-assisted rewrites, targeting “1 engineer, 1 month, 1 million lines of code.” The announcement followed acknowledgment that Windows 11 core components malfunctioned from July through October 2025, affecting Start Menu, Taskbar, File Explorer, and System Settings due to XAML framework complications.
When: Hunt announced the elimination plan on November 25, 2025, four months after Windows 11 problems began in July 2025. Microsoft officially acknowledged component malfunctions in late November 2025 and promised a December 2025 patch. The company established 2030 as the target date for completing C and C++ elimination across all products.
Where: The initiative operates within Microsoft’s Future of Scalable Software Engineering group in the EngHorizons organization under Microsoft CoreAI. Hunt opened a Principal Software Engineer position requiring in-person work in Redmond, Washington. Windows 11 malfunctions affected users globally across enterprise and consumer installations.
Why: Microsoft aims to eliminate technical debt at scale while pursuing “1 engineer, 1 month, 1 million lines of code” productivity through AI-assisted development. The strategy combines algorithmic infrastructure creating scalable graphs over source code with AI processing infrastructure applying agents guided by algorithms. The approach reflects broader industry movement toward memory-safe languages like Rust, though the initiative emerged after months of system failures potentially connected to AI-generated code quality issues.
