Published 3 days ago • loading... • Updated 10 hours ago

DeepMind Unveils AlphaEvolve, a Breakthrough AI for Algorithm Discovery

Google DeepMind unveiled AlphaEvolve, a new AI coding agent designed for algorithm discovery, on 14 May 2025 in a preprint release.
AlphaEvolve builds on prior DeepMind projects like AlphaTensor and AlphaDev, advancing a line of AI systems targeting algorithmic improvements in math and computing.
AlphaEvolve uses an evolutionary approach combined with Gemini language models to produce and evaluate complex code, making it more capable than earlier AI systems like FunSearch.
The system freed up 0.7% of Google’s global computing resources and reduced AI training time by 1%, with DeepMind calling it a “super coding agent” that has had substantial impact.
AlphaEvolve’s deployment suggests broader AI roles in optimizing infrastructure and tackling scientific problems, with DeepMind planning early academic access ahead of wider rollout.

Insights by Ground AI

Does this summary seem wrong?

31 Articles

All

Left

Center

Right

20minutos

Center

Google has succeeded: its AI can now create its own software without human help, goodbye to programmers

Mountain Viewers have followed the example of Meta and Microsoft and start working on AI agents to create the code of Google products and services.

2 days ago·Madrid, Spain

Read Full Article

The Hindu

Lean Left

Google DeepMind unveils AlphaEvolve, an AI agent that can generate new algorithms

Google DeepMind has announced a new AI agent that can solve complex coding and math problems, called AlphaEvolve.

2 days ago·India

Read Full Article

The Register

Center

Google DeepMind debuts algorithm evolving agent, AlphaEvolve

: AlphaEvolve may optimize your code in ways you hadn’t thought possible. Or not. Not is possible, too

2 days ago

Read Full Article

Ars Technica

Center

Google DeepMind creates super-advanced AI that can invent new algorithms

AlphaEvolve has already made Google’s data centers more efficient and improved Tensor chips.

2 days ago·United States

Read Full Article

TNW

Lean Left

5 impressive feats of DeepMind’s new self-evolving AI coding agent

AlphaEvolve makes large language models (LLMs) like Gemini better at solving complex computing and mathematical problems.

2 days ago·Amsterdam, Netherlands

Read Full Article

Wired

Reposted by

technewstube.com

Lean Left

Google DeepMind’s AI Agent Dreams Up Algorithms Beyond Human Expertise

A new system that combines Gemini’s coding abilities with an evolutionary approach improves datacenter scheduling and chip design, and fine-tunes large language models.

3 days ago·United States

Read Full Article

Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/year