Google Unveils AlphaEvolve, AI Agent That Invents New Algorithms

By Netvora Tech News

---

Google DeepMind has pulled back the curtain on AlphaEvolve, an artificial intelligence agent capable of inventing brand-new computer algorithms and deploying them across the company's vast computing empire. This groundbreaking technology pairs Google's Gemini large language models with an evolutionary approach that tests, refines, and improves algorithms automatically.

AlphaEvolve has already been deployed across Google's data centers, chip designs, and AI training systems, resulting in a significant boost in efficiency and solving complex mathematical problems that have stumped researchers for decades.

"AlphaEvolve is a Gemini-powered AI coding agent that is able to make new discoveries in computing and mathematics," explained Matej Balog, a researcher at Google DeepMind. "It can discover algorithms of remarkable complexity – spanning hundreds of lines of code with sophisticated logical structures that go far beyond simple functions."

The system represents a major leap in AI's ability to develop sophisticated algorithms for both scientific challenges and everyday computing problems. It dramatically expands upon Google's previous work with FunSearch, evolving entire codebases rather than single functions.

AI-Crafted Algorithms Power Google's Data Centers

AlphaEvolve's efficiency boost is a testament to its impact on Google's data centers. The AI agent has improved the company's computing efficiency by 0.7%, a significant gain that translates to real-world savings.

Breaking New Ground: AI Solves Matrix Multiplication Record

In another impressive feat, AlphaEvolve has solved the matrix multiplication problem, a challenge that has stood for 56 years. The AI agent has outperformed human researchers, achieving a breakthrough in a field that was thought to be the domain of human ingenuity.