Consensus is an AI-powered academic search engine built on a database of over 220 million peer-reviewed research papers. Unlike general AI tools, every response is tied back to a real research paper and is grounded in scientific research.

Over 5 million people already rely on Consensus to speed up their research and improve their overall experience. For many, it’s a more efficient and helpful alternative to traditional tools like PubMed and Google Scholar.

You can search using keywords or ask natural language questions. Our custom-built language models will then identify the most relevant papers and generate a clear summary - highlighting key insights from individual studies and across the research topic.

In this article, we explore why people choose Consensus and how they use it. Keep reading to learn more.

Most new AI tools, like ChatGPT, are fully generative, meaning the AI models are simply just generating text in response to a prompt. This can create a whole host of problems when doing academic research, namely an inability to reference actual research papers, or worse, making up fake papers.

Instead, Consensus begins its response with a comprehensive search of the literature to find relevant papers given the users input. Once a set of papers are returned, our AI models begin their work, extracting relevant insights from individual papers and summarizing takeaways across multiple papers.

Every piece of text that is generated by our AI models is tied back to its original source material with a citation. You are always one click away from the underlying source content.

Simply put, ChatGPT and other fully generative tools are not built for academic research. Consensus is.

Consensus contains a litany of features that are designed to both speed up your research, but also help guide you toward the highest quality research possible.

Some of our most popular features include:

Our goal is to make it easier to find high-quality research and want to use as many different ways to assess research quality, beyond just standard bibliographic metrics.

Consensus uses AI to help make research more efficient, accessible, and effective. We also know powerful these models are and how they must be used carefully and responsibly.

We have many guardrails in place to help ensure responsible AI usage in our product:

By state-of-the-art using natural language processing methods, Consensus is designed to take more flexible queries than traditional academic search engines to make searching feel even more intuitive.

Instead of trying to hack together keywords, try a few of the best ways to search on Consensus:

🔍 Simple yes/no question: does the death penalty reduce crime?

🔍 Open-ended phrase: benefits of mindfulness meditation

🔍 Ask about the effects of a concept: what is the impact of climate change on GDP?

🔍 A conversational command about a research topic: review the literature about the relationship between socioeconomic status and healthcare outcomes

Instead of delivering you a list of links to papers related to your search, leaving you to click open dozens of tabs, Consensus will pull out direct insights from papers related to your question.

This allows you to more quickly find the papers to dig into further, or in some cases, deliver you the insight you were looking for right on the search results page.

Additionally, with Pro Analysis, you can instantly get an overview of what the research says about your query across many papers all at once.

With Pro, you can even ask for specifics formats of the synthesis of the research you are looking for. Try the search below:

The most straightforward use case for Consensus. You are writing a research paper and need to find supporting evidence for your hypothesis.

You type in your research question into Consensus and will instantly see where certain papers stand in regards to your hypothesis, allowing you to quickly decide which papers to further investigate and eventually cite in your work.

Similar to the above, the only difference being that you come to Consensus without a hypothesis in mind.

You begin by typing your research question into Consensus, quickly analyze the weight of evidence on both sides of the argument and make a determination of where you stand and what the argument your work will be taking.

If you are tasked with a literature review, Consensus can be an easy way to find papers related to multiple facets of your research question.

For example, let’s say you are doing a literature review on the effects of climate change on the global economy. You can first start with a question like “What is the impact of climate change on the economy?” Then, get more granular with questions like “How will climate change affect GDP?” and “Will climate change have a more negative impact on developing countries?”

What if you don’t even have a question in mind? Well, Consensus can still be helpful.

If you are writing a paper, type a potential research question into Consensus to see where the literature stands and what would be best to research further.

Or, if you are conducting research yourself and are looking for an unexplored topic, ask some questions in Consensus to see if you have identified a gap in the literature.

Need to know what researchers think about your question? Just type your question into Consensus and see conclusions from relevant research instantly.

If you need to dig deeper, click into the details page to read the abstract or access the full text, but Consensus is the easiest way to get a snapshot of the research in seconds.