Microsoft's 'North Star': The 2030 Initiative to Eradicate C/C++ via AI Automation
A research project targeting a '1-1-1' efficiency metric aims to transition legacy systems to Rust, though immediate Windows rewrites remain off the table.
In a move that signals a potential paradigm shift for enterprise software engineering, Microsoft Distinguished Engineer Galen Hunt has unveiled an ambitious research initiative aimed at eliminating C and C++ from the company's codebase by 2030. Revealed through a recruitment notice posted to LinkedIn in late December 2025, the project seeks to replace legacy code with memory-safe Rust using a highly automated AI infrastructure. While Hunt has since clarified that this represents a 'North Star' research capability rather than an immediate product roadmap for Windows, the initiative's target metric-enabling a single engineer to migrate one million lines of code in one month-has ignited intense debate regarding the feasibility, safety, and future role of human oversight in software development.
The core of this initiative, described by Hunt as the 'Future of Scalable Software Engineering,' centers on a radical increase in developer velocity. The stated objective is to develop the operational capacity to 'eliminate every line of C and C++ from Microsoft by 2030'. To achieve this, the team is targeting a '1-1-1' efficiency metric: '1 engineer, 1 month, 1 million lines of code'. This scale of refactoring is mathematically impossible under current manual workflows, necessitating a heavy reliance on autonomous systems.
According to the technical details released, the approach moves beyond simple Large Language Model (LLM) translation. Instead, Microsoft intends to deploy a 'powerful code processing infrastructure' where algorithms generate a 'scalable graph over source code'. These graphs provide the structural logic and dependency mapping required to ground 'AI agents', which then perform the actual translation and rewriting tasks. This hybrid model-combining deterministic algorithms for topology with probabilistic AI for syntax translation-aims to mitigate the hallucination risks inherent in pure generative AI approaches.
The Driver: Memory Safety and Technical Debt
The push toward Rust is driven by an industry-wide mandate to eradicate memory safety vulnerabilities, a class of bugs responsible for approximately 70% of historical security patches in Microsoft products. While the White House Office of the National Cyber Director (ONCD) has urged the industry to abandon C and C++, the sheer volume of legacy code-tens of millions of lines within the Windows kernel alone-has previously made total migration cost-prohibitive. Microsoft's hypothesis is that AI automation is the only lever capable of overcoming this economic barrier.
Clarification vs. Reality
Following the viral spread of the job posting, Hunt issued a clarification on December 24, 2025, emphasizing that the 2030 target is a research vision to build the capability for such a transition, not a confirmed timeline for rewriting Windows 11 or 12. He explicitly stated, 'Windows is NOT being rewritten in Rust with AI' in the immediate term. However, the existence of the team suggests that Microsoft views automated legacy migration as an inevitability, provided the tooling can mature sufficiently.
Industry Skepticism: The 'Black Box' Risk
The proposal has drawn sharp criticism from senior software engineers and industry observers. The primary concern is the 'human review bottleneck.' If an AI agent generates code at a rate of one million lines per month, a human engineer would need to review approximately 800 lines per minute to maintain oversight. Critics argue this effectively removes human auditing from the loop, replacing 'battle-tested' legacy code with 'untested AI-generated material'.
Furthermore, technical hurdles regarding the Win32 API remain significant. Much of the Windows kernel relies on direct memory manipulation that Rust classifies as 'Unsafe.' Automating the translation of C++ to Rust while wrapping these interactions in 'Unsafe' blocks could theoretically preserve the code's function but negate the memory safety benefits the project aims to achieve.
Ultimately, this initiative represents a high-stakes experiment in the evolution of software engineering. If successful, it transitions the engineer's role from a 'producer' of code to an 'architect' and 'auditor' of AI systems. If it fails, it highlights the stubborn resilience of legacy infrastructure against the promises of generative automation.
Key Takeaways
- The '1-1-1' Metric: Microsoft is targeting a research capability allowing one engineer to process one million lines of code in one month.
- Hybrid AI Architecture: The approach utilizes algorithms to build structural graphs of the codebase, which then guide AI agents in translation, aiming to reduce errors common in pure LLM approaches.
- Strategic Goal: The 'North Star' vision is the total elimination of C and C++ by 2030 to resolve memory safety vulnerabilities, though this is not yet a confirmed product roadmap.
- Quality Assurance Risks: Critics note that the targeted velocity exceeds human cognitive limits for code review, raising concerns about the stability of AI-generated systems.
- Legacy Compatibility: The project faces significant technical challenges in mapping complex, unsafe Win32 APIs to Rust without losing the safety guarantees of the new language.