Elicit - Automating Research
Einstein said the important thing is to never stop questioning, a truth that fuels scientific discovery, the heartbeat of progress. The potential for AI to positively shape the future is perhaps greatest in accelerating research itself – enhancing our ability to ask better questions, find relevant information faster, and draw connections that might otherwise remain hidden.
Elicit is building an AI Research Assistant that is used by over 400,000 researchers each month to accelerate their research and automate time-consuming tasks like summarizing papers, extracting data, and synthesizing findings.
Today, Elicit is introducing new features that use AI to automatically generate Systematic Reviews and Research Reports, delivering in minutes what has traditionally taken months of manual work. For a given research question, Elicit searches for potentially relevant papers from a corpus of 125 million, writes its own screening criteria and filters the studies based on the inclusion criteria, extracts key metrics and data from the filtered set of papers, and then synthesizes the findings into a report. All automatically via AI.
Automating the ability to understand what is known is just the first step towards Elicit’s mission to radically increase good reasoning in the world. The plan is to help automate discovery and the scientific method while generalizing scientific rigor to many more domains.
The future envisioned by Elicit's co-founders, Andreas Stuhlmüller and Jungwon Byun, sees Elicit being used by medical researchers to quickly synthesize findings across thousands of studies to identify promising treatments. By policy makers who rapidly assess the evidence base for interventions. By scientists who can navigate vast literature landscapes to discover overlooked connections and accelerate breakthroughs. By removing the friction from the research process, and guiding humans to better decisions, Elicit transforms how knowledge accumulates and flows, turning what was once a painstaking process of discovery into something that unfolds in real time, with positive implications for human progress.
To bring this vision to reality, Andreas and Jungwon have built a talented team across engineering, machine learning, and product design. We were impressed by the quality of everything Elicit does - from their product to their job descriptions, the team ships quality, quickly.
Spark is proud to support Andreas, Jungwon, and the team at Elicit as they work to automate research and build the infrastructure for good reasoning at scale. Together with Footwork we are co-leading a $22M investment in Elicit joined by Fifty Years, Basis Set, and Mythos.
With this funding Elicit will expand beyond academic research, bringing systematic, evidence-based AI decision-making to organizations across healthcare, policy-making, scientific research, and many other domains. In a world increasingly shaped by artificial intelligence, Elicit’s approach – using AI to help radically improve reasoning – represents the kind of thoughtful innovation we need. We’re excited to join Elicit on this journey.