Claude Agents Find a New Enzyme System With CRISPR-Like Repeats
Anthropic said on September 23, 2026 that Claude agents found ART, a phage enzyme system with CRISPR-like repeats, as it unveiled a new life sciences lab.
Research & Science
AI research news: breakthroughs in reasoning, mathematics, science, biology and robotics, plus major studies on AI capabilities and productivity.
8 stories
Today's research becomes tomorrow's product features. This category highlights significant AI research results, from mathematical reasoning and scientific discovery to world models and robotics. It also covers studies that measure how AI changes the way developers work.
Anthropic Study: Domain Expertise Predicts Success With Coding Agents
Anthropic's June 16, 2026 study of 400,000 Claude Code sessions found domain expertise, not a coding background, best predicts success with coding agents.
Google DeepMind Publishes Co-Scientist in Nature and Opens It to Researchers
Google DeepMind published its Co-Scientist multi-agent research system in Nature in May 2026 and began offering it to scientists through Gemini for Science.
Google DeepMind Unveils Genie 3, a Real-Time Interactive World Model
Google DeepMind announced Genie 3 in August 2025, a world model that turns text prompts into interactive 720p worlds at 24 fps that stay consistent for minutes.
METR Study Finds Early-2025 AI Tools Slowed Experienced Developers by 19%
A July 2025 METR randomized trial found experienced open-source developers took 19% longer with early-2025 AI tools, though they believed AI made them faster.
Google DeepMind Unveils AlphaGenome, a Unified Model for DNA's Regulatory Code
Google DeepMind introduced AlphaGenome in June 2025, a model that reads up to 1 million DNA base pairs to predict how genetic variants affect gene regulation.
Google DeepMind Unveils AlphaEvolve, a Gemini-Powered Agent That Discovers Algorithms
Google DeepMind introduced AlphaEvolve in May 2025, a Gemini-powered evolutionary coding agent that found faster algorithms and saves 0.7% of Google's compute.
Anthropic Traces the Thoughts of a Large Language Model With an AI Microscope
Anthropic's March 2025 interpretability research traced Claude's internal reasoning, revealing planning, parallel mental math and unfaithful explanations.
No matches. Try the site search.