G5 Labs, a spinout from MIT’s Computer Science and Artificial Intelligence Laboratory, has emerged from stealth with $14 million in seed funding to develop a new abstraction layer for AI-native software development that treats natural language as source code.
The round was co-led by Pillar VC and Battery Ventures, with participation from Omega Venture Partners and Encoded Ventures. Angel investors include Jeff Dean, CEO of Discovery Loop and former Chief Scientist at Google.
G5 Labs is addressing a problem emerging as AI coding tools dramatically increase the amount of software individual developers can produce. While developers can now generate large volumes of code quickly, organizations still need to review, merge, secure, govern and maintain that software.
Rather than simply adding another AI coding assistant, G5 is building an abstraction layer above existing development tools.
Its platform represents natural-language business and technical intent as a structured ontology graph. That intent effectively becomes the source code from which software can be generated, reviewed and governed.
G5 emerged from MIT CSAIL research centered on a self-learning, bi-directional compiler capable of converting natural language into code and code back into structured natural-language intent.
The approach allows organizations to manage software intent across an entire company, resolve merge conflicts at the semantic level instead of the code level and trace individual lines of code back to the requirements that produced them.
G5 also aims to help organizations control AI development costs before tokens are consumed and enforce policies involving areas such as GDPR and cybersecurity across both legacy and newly created systems.
That architecture could also allow more people outside traditional engineering teams to participate directly in software development.
Product owners, business analysts and compliance teams can define requirements and governance rules using familiar business language rather than depending entirely on developers to translate those requirements into code.
One early deployment involved modernization of a financial services application. G5 said its ontology-level comparison identified structural problems that may have been missed during a conventional migration from an older application to a new system.
By presenting conflicts in both technical and business-oriented formats, the platform allowed business analysts to participate directly in design decisions and accelerate implementation of new functionality.
G5 is initially working with organizations modernizing complex software in regulated industries.
Instead of simply translating legacy code into a newer programming language or framework, the company is attempting to rebuild systems around current business intent while avoiding the technical debt and outdated logic accumulated in older applications.
The result is intended to make modernized systems easier to govern and better suited for future AI-driven development.
The broader vision is that software development could eventually move away from programming languages such as Java, C, Python and Rust as the primary interface for creating applications, with structured natural language becoming a higher-level abstraction.
G5 Labs was founded by a team that includes MIT PhDs and veterans of Google and AWS. CEO Tim Kraska is a professor at MIT CSAIL and co-director of MIT’s Generative AI Impact Consortium. He has spent roughly two decades working on machine learning and AI across academia, Google, Amazon and two previous startups that were acquired.
The company will use the $14 million round to expand its engineering organization, scale customer deployments and continue developing the G5 platform and ontology compiler.
KEY QUOTES:
“Teams are either struggling to adapt to AI or drowning in AI-generated code they can’t review, merge, or trust. Single developers generate tens of thousands of lines a day, maintainers are overwhelmed with AI slop, security teams face entirely new threats, and businesses are racking up six-figure monthly AI bills with little to show for it. The fundamental problem is that we have slapped AI onto development tools and processes designed for humans. This isn’t just better prompting — it’s a new abstraction layer. With G5, natural-language intent becomes the source code itself: compilable, mergeable, diffable, and governable, the way high-level languages once replaced assembly. For the first time, it puts software in the hands of the whole organization, not just developers.”
Tim Kraska, Co-Founder and CEO of G5 Labs and Professor at MIT CSAIL
“Past major leaps in software have always come from raising the level of abstraction. G5 is a serious, well-engineered attempt at the next one.”
Jeff Dean, CEO of Discovery Loop and Former Chief Scientist at Google
“G5’s approach is the first credible answer to the AI productivity paradox we’re seeing across complex systems. The team’s research pedigree and early enterprise traction convinced us this is the company that defines the category.”
Parker McKee, Partner at Pillar VC
“With AI, there will soon be no more reason to write software in Java, C, Python, or Rust. The new language of software development will be English. The missing link is a suite of tools to build a robust software-engineering process around it. G5 is delivering those tools.”
Max Schireson, Partner at Battery Ventures and Former CEO of MongoDB
