Execute and evaluate a configured chemical experiment
See the requirement-level assessment and its sources.
SCIENCE UNDER AUTOMATION
AI in Chemical sciences: capabilities and evidence
Chemistry couples molecular proposals to synthesis, measurement and interpretation. A configured robotic loop can cover much of one experiment.
Autonomy · up to 5.5 / 10
Model interval: 3.2–5.5. Assessed 21 October 2026.
Two equally weighted workflow families: a bounded operational task and an investigation or deployment beyond that task. Inherited subfields use this parent scenario.
Calculation and work mix
Each requirement uses the operational scale from 0 (independent execution unestablished) to 4 (reliable independent execution in the stated setting). A workflow takes its lowest requirement level. The two workflows receive equal weight; the headline is 1 + 9 × weighted level / 4.
These levels are source-linked model assignments, not measurements reported by the papers. “Up to” is the upper model bound, not a probability or a forecast date.
Execute and evaluate a configured chemical experiment: 2–3 / 4.
Establish a new reaction or chemical explanation: 0–1 / 4.
Upper bound with 75:25 workflow weights: 6.6. With 25:75 weights: 4.4.
What has AI demonstrated?
Twenty agent configurations were evaluated on five chemical protocol tasks. Strong configurations exceeded 0.89 procedural F1 on challenging multi-plate synthesis tasks.
- Test
- Peer-reviewed protocol evaluation with human-intervention conditions
- Comparison
- Comparators vary by study; see the linked sources.
- System and version
- AutoLabs · 2026-06-25
- Evidence checked
- 2026-10-21
The output is a procedure for configured liquid-handling hardware. Procedural F1 is distinct from experimental discovery or a complete autonomous laboratory.
What does the field require?
Execute and evaluate a configured chemical experiment
- Translate the goal into an executable protocol
Autonomy level: 2–3 / 4.
AutoLabs approaches expert procedure references with strong reasoning and self-correction on its five tasks.
AutoLabs - Run synthesis and acquire complementary measurements
Autonomy level: 2–3 / 4.
Mobile robots link synthesis, NMR and mass spectrometry in a configured laboratory.
Autonomous mobile robots for exploratory synthetic chemistry - Choose follow-up experiments within defined criteria
Autonomy level: 2–3 / 4.
The closed loop selects and repeats hits using scientist-authored criteria.
Autonomous mobile robots for exploratory synthetic chemistry
Establish a new reaction or chemical explanation
- Select a defensible reaction space
Autonomy level: 0–1 / 4.
Domain experts selected reactions and building blocks in the physical demonstrations.
Autonomous mobile robots for exploratory synthetic chemistry - Configure equipment for new conditions
Autonomy level: 1–2 / 4.
Procedure generation targets supported hardware; new configurations require experimental expertise.
AutoLabs - Resolve unexpected chemistry with independent checks
Autonomy level: 1–2 / 4.
Multiple instruments support verification, but heuristics and interpretation criteria remain authored by experts.
Autonomous mobile robots for exploratory synthetic chemistry
Research tasks
Establish a new reaction or chemical explanation
See the requirement-level assessment and its sources.
Can AI replace the whole workflow?
The results above measure specific tasks. The field assessment also requires completion of the connected workflow, including the requirements listed here. A field-wide replacement percentage has not been established.
Read the assessment method