Google DeepMind BlogMay 12, 2026, 2:40 PMimportant 85

Google DeepMind 推出 Co-Scientist:基於 Gemini 的多 Agent 科學研究 AI 助手

Original: Co-Scientist: A multi-agent AI partner to accelerate research

Google DeepMind has officially launched "Co-Scientist," a new multi-agent AI collaboration system built on the Gemini model, designed to…

Google DeepMind 發表全新 AI 系統「Co-Scientist」,採用基於 Gemini 的多 Agent(多智慧體)架構。該系統能扮演科學家的虛擬合作夥伴,協助進行文獻回顧、提出新穎假設、設計實驗步驟並分析複雜數據。透過不同專業 Agent 的協作與互相審查,Co-Scientist 旨在大幅縮短科研週期,推動生物、化學及材料科學等領域的突破。

Google DeepMind has officially launched "Co-Scientist," a new multi-agent AI collaboration system built on the Gemini model, designed to serve as a virtual collaborator for researchers and comprehensively accelerate scientific discovery and technological breakthroughs. As scientific knowledge continues to grow explosively, it has become nearly impossible for any single researcher to keep up with the latest advances across all fields. Co-Scientist was created specifically to address the highly complex and time-consuming processes in scientific research — namely literature review, hypothesis generation, experimental design, and data analysis.

Full summary

Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.

See Pro plans →

Want the original English / full article?

Read on Google DeepMind Blog →

Summaries are AI-generated; the original article is authoritative.