THE LAB
SECTION 11
ISSUE 001
Sequence-to-Experiment Compilers
Projection: the useful interface for biological AI will look less like a chatbot than a compiler. It will turn a sequence hypothesis into controls, materials, assays, stopping rules, and a provenance trail—then refuse to compile when uncertainty or dual-use risk crosses the laboratory’s approved boundary.
Why this idea is here
What the evidence establishes.
Genome models generate coherent candidate sequences and report experimentally validated guided designs; autonomous-instrument research presents a human-in-the-loop pipeline with optional expert input for complex multimodal scientific workflows. The compiler and safety envelope are the labeled projection.
Source ledger
Read the sources.
- S01Genome Modelling and Design Across All Domains of Life with Evo 2
peer-reviewed primary research / dated 2026 / retrieved 2026-07-09
- S02Operating Advanced Scientific Instruments with AI Agents That Learn on the Job
peer-reviewed primary research / dated 2026 / retrieved 2026-07-09