Elicit

Elicit operates as an AI-powered research assistant, developed to automate and scale complex reasoning and research workflows. The company originated as a project within Ought, a non-profit machine learning research lab, before spinning off as an independent public benefit corporation in 2023. This transition was supported by a $9 million seed funding round in September 2023, led by Fifty Years, and was followed by a $22 million Series A in February 2025, co-led by Spark Capital and Footwork.

The company was co-founded by CEO Andreas Stuhlmüller and COO Jungwon Byun. Stuhlmüller's background is rooted in cognitive science and machine learning, holding a Ph.D. from MIT and having conducted postdoctoral research at Stanford, where he focused on automating reasoning. Byun brings experience in growth and operations, having previously served as Head of Growth at Upstart and as a consultant at Oliver Wyman, with a degree in economics from Yale. Their combined expertise reflects the company's mission to merge machine learning with practical, scalable business applications. The initial concept for Elicit grew out of Ought, which the founders established to explore how machine learning could assist with deliberation and open-ended reasoning.

Elicit's core product is a platform designed to streamline the research process, primarily for academics, scientists, and analysts at research-focused organizations. It utilizes large language models to automate tasks such as literature reviews, data extraction, and evidence synthesis. The platform searches a database of over 125 million academic papers to find relevant studies, summarize key takeaways, and extract specific data points into structured formats. One of its key features is the ability to break down complex research questions into smaller, manageable subtasks that the AI can execute more reliably, a method described as supervising the process rather than just the outcome. This approach is intended to minimize inaccuracies and ensure the transparency of results, as all extracted information is linked back to the source text.

The business model is subscription-based, with tiered plans aimed at individual researchers, academic institutions, and corporate R&D teams. By automating time-consuming aspects of research, Elicit aims to significantly increase the productivity and accuracy of its users, enabling them to focus on higher-level analysis and decision-making. The platform is expanding beyond academic literature to serve a broader range of evidence-based decision-making in various industries.

Keywords: AI research assistant, literature review automation, systematic reviews, evidence synthesis, data extraction, machine learning for research, academic research tools, scientific workflow automation, natural language processing, research productivity, Andreas Stuhlmüller, Jungwon Byun, Ought, public benefit corporation, cognitive science AI, computational cognitive science, research automation platform, evidence-based decision making, AI for science, knowledge work automation