How scientists use AI for hypothesis generation, analysis, literature review and experimental design.
AI accelerates scientific discovery across every discipline — from generating novel hypotheses to analysing data at unprecedented scale.
The key challenge is maintaining rigor: hypothesis testing, peer review and reproducibility remain firmly in human hands.
Elicit and Consensus synthesise findings across thousands of papers, identifying consensus, contradictions and gaps. Systematic reviews in days rather than months.
AI analyses existing data to suggest novel hypotheses. It designs experiments optimising sample sizes and statistical approaches.
AI processes complex, high-dimensional datasets — genomic, imaging, climate — identifying patterns invisible to traditional methods.
AI generates hypotheses but credibility requires rigorous design, peer review and reproducibility — all human processes.
Yes — literature synthesis, significance statements and formatting. The core vision must come from the researcher.
Most require disclosure of AI usage. AI is accepted as a tool but not as a co-author.
Ready to learn AI by building real projects?