NASA's latest mission to the ISS features a bacterial experiment

Scientists are sending several strains of disease-causing bacteria to the International Space Station as part of the Crew-11 mission. This experiment isn't the plot to some cheesy horror film, bu [...]

Alibaba's model never trained as an agent — and improved agent performance across seven benchmarks

Alibaba's Qwen team released Qwen-AgentWorld on Tuesday — two models trained not to act inside agent environments, but to predict what those environments return. The release covers seven domain [...]

Cockroaches scurry around with thousands of pieces of bacterial genomes

Transferring genes across species doesn't just happen in microbes. [...]

Databricks built a RAG agent it says can handle every kind of enterprise search

Most enterprise RAG pipelines are optimized for one search behavior. They fail silently on the others. A model trained to synthesize cross-document reports handles constraint-driven entity search poor [...]

Intent-based chaos testing is designed for when AI behaves confidently — and wrongly

Here is a scenario that should concern every enterprise architect shipping autonomous AI systems right now: An observability agent is running in production. Its job is to detect infrastructure anomali [...]

Neanderthals seemed to have a thing for modern human women

"Neanderthal deserts" in our genomes suggest a strong pattern in matings. [...]

Generare raises €20M to decode the 97% of microbial chemistry

The Paris techbio company screens microbial genomes to find molecules that evolution spent three billion years producing, and claims to have characterised more novel small molecules in 2025 than the r [...]

Scientific AI models trained on different data are learning the same internal picture of matter, study finds

MIT researchers compared 59 scientific AI models and found they converge to similar internal representations of molecules, materials, and proteins, even when built on different architectures and train [...]

New 'Markovian Thinking' technique unlocks a path to million-token AI reasoning

Researchers at Mila have proposed a new technique that makes large language models (LLMs) vastly more efficient when performing complex reasoning. Called Markovian Thinking, the approach allows LLMs t [...]