Consensus’ cover photo
Consensus

Consensus

Technology, Information and Media

San Francisco, CA 22,533 followers

AI search engine for research papers

About us

Consensus is an AI-powered search engine designed to make searching for and consuming scientific research papers easier. Our product was launched in the winter of 2022 and we currently serve millions of students, researchers, doctors and consumers alike from across the world.

Website
https://consensus.app/
Industry
Technology, Information and Media
Company size
2-10 employees
Headquarters
San Francisco, CA
Type
Privately Held
Founded
2022

Locations

Employees at Consensus

Updates

  • Consensus reposted this

    ⚕️ How one doctor used Consensus to uncover life-saving evidence... It’s easy to get swept up in the negative headlines about AI. There’s plenty to debate, plenty to critique, and plenty we need to get right. But it’s also worth remembering that some tools are already making a real, tangible difference. Not tomorrow, today. I’m incredibly grateful to Dr. Multani for allowing us to share his story of exactly this in action. A patient at an Iowa hospital arrived with an infected knee and a rare blood disorder. An unusual combination with no clear treatment guideline. Faced with uncertainty and a ticking clock, Dr. Multani turned to Consensus and quickly surfaced an obscure case report from Japan describing a nearly identical situation. That evidence helped guide a life-saving treatment plan for this patient. Moments like this remind us why we’re building Consensus in the first place: to help real people make better, faster decisions using the best evidence possible. It’s an honor to play even a small role in outcomes that truly matter 💜 Read our story here: https://lnkd.in/e2yYkdig

  • Consensus reposted this

    Few AI products have access to the full text of published research. That’s why we (Consensus) are thrilled to announce our partnership with American Chemical Society, the world’s leading publisher of chemistry research, to bring their trusted, peer-reviewed full-text content into Consensus. Through this partnership, ACS is enabling Consensus to access the full-text content of its paywalled journals. This allows Consensus to deliver more precise search results and contextually-accurate AI-generated previews drawn directly from the full-text of ACS’s authoritative research. Unlike general-purpose AI tools that train on publishers’ content, Consensus drives discovery of it. Our users click through at exceptional rates to verify, read, and cite the original sources—boosting visibility, discoverability, and long-term impact for our publishing partners. This makes this partnership truly a win-win for the entire ecosystem: For researchers (Consensus users), this means more relevant search results and more robust AI summaries grounded in full-text content. For publishers like ACS, it allows them reach a broader, next-generation audience of engaged researchers who are eager to go deeper into the original work. Huge shoutout to JL Needham, David Wallace, Martin Majlund, and Ralph Youngen for all of your tireless efforts on getting this partnership across the finish line. We’re honored to work with ACS to make trusted science more discoverable, understandable, and actionable for everyone 💪 Read more on our partnership here: https://lnkd.in/eT745uVV

  • Consensus reposted this

    So stoked to launch our the Consensus Scholar Agent with OpenAI today! No more juggling between Google Scholar searches and tabs, put your entire research search request into Consensus and let our agent do the rest 💫 Also, you know have hired a great team when your engineers are 10x better on camera than the CEO - Aakash Adesara KILLS it here at OpenAI Dev Day 🔥 🔥

    View organization page for OpenAI for Startups

    11,045 followers

    Most scientists spend more time reading papers than running experiments. Consensus is changing that. What started as a search engine for academic papers is now an agentic system that reads, reasons, and cites like a researcher — built with OpenAI's GPT-5 and the Responses API. Co-founders Christian Salem and Eric Olson built Consensus to make scientific knowledge more accessible and usable—because when researchers can trust evidence faster, discovery accelerates. 🎥 We caught up with Aakash Adesara at DevDay to hear how they built it — full story in comments ↓

  • Consensus reposted this

    Most scientists spend more time reading papers than running experiments. Consensus is changing that. What started as a search engine for academic papers is now an agentic system that reads, reasons, and cites like a researcher — built with OpenAI's GPT-5 and the Responses API. Co-founders Christian Salem and Eric Olson built Consensus to make scientific knowledge more accessible and usable—because when researchers can trust evidence faster, discovery accelerates. 🎥 We caught up with Aakash Adesara at DevDay to hear how they built it — full story in comments ↓

  • Today we launched Scholar Agent 🎓 🤖, the only AI agent purpose-built for academic literature review. Try it for free on Consensus! Thank you OpenAI for Startups for showcasing our mission to accelerate research 🔬 👩🔬

    View organization page for OpenAI for Startups

    11,045 followers

    Most scientists spend more time reading papers than running experiments. Consensus is changing that. What started as a search engine for academic papers is now an agentic system that reads, reasons, and cites like a researcher — built with OpenAI's GPT-5 and the Responses API. Co-founders Christian Salem and Eric Olson built Consensus to make scientific knowledge more accessible and usable—because when researchers can trust evidence faster, discovery accelerates. 🎥 We caught up with Aakash Adesara at DevDay to hear how they built it — full story in comments ↓

  • Consensus reposted this

    PhD Students - Stop Wasting Weeks on Literature Reviews. How to validate your research questions in 20 minutes instead of 4 weeks? Stop drowning in 147 browser tabs wondering if your question is even viable. The problem with traditional literature reviews: ↳ 3+ weeks just to validate if your research question is worth pursuing ↳ No clear way to compare evidence across different approaches ↳ Zero visibility into whether sufficient research exists Meet Consensus Scholar Agent. An AI copilot that validates your PhD research direction before you waste months on the wrong question. How Scholar Agent validates your research questions: 1️⃣ Enter your research question with smart filters to search across 220+ million papers 2️⃣ Review AI-generated analysis with instant summaries of key findings from top studies 3️⃣ Ask targeted follow-up questions to explore methodology, limitations, and research gaps 4️⃣ Check the Consensus Meter for visual scientific agreement (High/ Moderate/Low/Insufficient findings) 5️⃣ Export comprehensive PDFs with all findings and citations to share with your supervisor Why Scholar Agent works: ✅ Compares multiple approaches in one search ✅ Validates question viability in 5 minutes ✅ Provides citation-backed evidence instantly ✅ Shows research consensus visually ✅ Saves weeks of manual screening Pro tip: Use Scholar Agent in Week 1 of your PhD to validate your research direction early. The #1 dissertation killer? Starting with the wrong question and discovering it 6 months later. Stop guessing if your research question is valid. Get evidence-based validation in 20 minutes. 👉 Start with Consensus free, or unlock 1 month of Pro free with my link: https://lnkd.in/eykd7hvJ. Pro supercharges your research with more deep searches and unlimited Pro analyses.

    • No alternative text description for this image
  • Consensus reposted this

    🚢 Another HUGE release for Consensus this week! We are thrilled to be launching Medical Mode 🩺 Medical Mode focuses Consensus’ search and AI analysis exclusively on the highest-quality medical evidence: 8M papers + 50K clinical guidelines from the world’s top 1k medical journals. Consensus covers every field of science imaginable. This scope makes it an incredible tool for academics, professionals, and clinicians conducting literature reviews, finding references, or exploring complex questions 🔬 But when it comes to medicine, breadth isn’t always better and precision matters most. This is why we built Medical Mode. It is designed for clinicians who need focused, evidence-based answers from trusted sources. The journals for Medical Mode were selected using a combination of top ranking systems like: Impact Factor, SJR, and SciScore. It includes top journals like: NEJM, JAMA, BMJ, Cochrane Database of Systematic Reviews, Annals of Internal Medicine, Nature Medicine, and more. Switching to Medical Mode in Consensus is easy: 1️⃣ Click the new source dropdown on the left-hand side of the search bar 2️⃣ Select “Medical Mode” 3️⃣ Your interface turns green, indicating you’re now searching only clinical sources Medical Mode touched a lot hands at Consensus and it's always awesome to see so much cross-team collaboration: David Wallace kicked things off with his typical superhuman data ingestion and cleaning efforts to create the Medical Mode corpus. Yuri Fursov took the hand-off from David and made this corpus searchable and performant for our users. And then Alan Kalbfleisch finished things off and delivered on the web development work alongside his casual dozen or so other ongoing projects. Medicine moves fast. Evidence should move faster. Medical Mode is built for questions where evidence quality can’t be compromised. Join the millions of researchers and clinicians, try Medical Mode in Consensus today: https://lnkd.in/euJZt4fH

    • No alternative text description for this image
  • Consensus reposted this

    🚢 Another HUGE release for Consensus this week! We are thrilled to be launching Medical Mode 🩺 Medical Mode focuses Consensus’ search and AI analysis exclusively on the highest-quality medical evidence: 8M papers + 50K clinical guidelines from the world’s top 1k medical journals. Consensus covers every field of science imaginable. This scope makes it an incredible tool for academics, professionals, and clinicians conducting literature reviews, finding references, or exploring complex questions 🔬 But when it comes to medicine, breadth isn’t always better and precision matters most. This is why we built Medical Mode. It is designed for clinicians who need focused, evidence-based answers from trusted sources. The journals for Medical Mode were selected using a combination of top ranking systems like: Impact Factor, SJR, and SciScore. It includes top journals like: NEJM, JAMA, BMJ, Cochrane Database of Systematic Reviews, Annals of Internal Medicine, Nature Medicine, and more. Switching to Medical Mode in Consensus is easy: 1️⃣ Click the new source dropdown on the left-hand side of the search bar 2️⃣ Select “Medical Mode” 3️⃣ Your interface turns green, indicating you’re now searching only clinical sources Medical Mode touched a lot hands at Consensus and it's always awesome to see so much cross-team collaboration: David Wallace kicked things off with his typical superhuman data ingestion and cleaning efforts to create the Medical Mode corpus. Yuri Fursov took the hand-off from David and made this corpus searchable and performant for our users. And then Alan Kalbfleisch finished things off and delivered on the web development work alongside his casual dozen or so other ongoing projects. Medicine moves fast. Evidence should move faster. Medical Mode is built for questions where evidence quality can’t be compromised. Join the millions of researchers and clinicians, try Medical Mode in Consensus today: https://lnkd.in/euJZt4fH

    • No alternative text description for this image
  • We’re excited to launch one of our most-requested features: the ability to select specific papers (up to 5 unique papers at a time) for AI analysis 🎉 You can now get comprehensive, cited AI analysis of full-text papers by either: asking a follow-up of selected papers or uploading your own PDFs. Use the new drop-down menu on the search bar to utilize this new functionality!

    🚨 New in Consensus this week: you can now select up to five specific papers for detailed AI analysis. This was one of our most-requested features, and it’s a big one. You can now get comprehensive, cited AI analysis of full-text papers by either: - Asking a follow-up of selected papers - Or uploading your own PDFs 📜 To chat with uploaded papers, click the new drop down menu on the left-hand side of the search bar to upload papers and select from previously uploaded papers for detailed full-text analysis. ❓ To direct your follow-up questions at specific papers, scroll to your list of search results and click the new checkbox next to each paper. You can select up to five papers for your follow-up query. This new functionality helps you dig in where it matters most. Move beyond surface-level summaries to see how studies align, where they differ, and what their methods and limitations reveal about the bigger picture. Great research lives in the details. 👏 Massive shoutout to Jimmy Longley for driving this from conception to completion. This project touched nearly every product surface, and Jimmy knows them all inside out. As always, he navigated it swiftly and with diligence. Pablo Garcés also deserves a big call out for setting the vision of how this detailed analysis functionality could fit within the core Consensus search & analysis workflow (RIP Ask Paper 🪦 ). Many more updates are on the way - stay tuned! 🏗️

Similar pages

Browse jobs

Funding