Today, we’re taking the next step toward our mission with the launch of the Elicit Research Agent, an AI environment built for high-stakes decisions. You can use the Research Agent today at elicit.com or access it through API (docs.elicit.com) to bring its capabilities into your own workflows.

The future of agentic AI depends on trustworthy information infrastructure. AI agents can search, reason, and generate answers, but need the right evidence to do so well. That is why we built the Elicit API and MCP server. Watch the demo recording: youtu.be/MSwk9DEKpYA

AI Agents Grounded in Evidence: A Live Demo of the Elicit API and MCP Server

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

youtube.com

We are one day away from a live demo of the Elicit API and MCP server. We'll cover the Search API, Reports API, Systematic Review API, and grounding Claude and ChatGPT with Elicit. Register to attend live or get a copy of the recording: lnkd.in/g47XBm8x

Join us for a live demo of the Elicit API and MCP server. Title: AI Agents Grounded in Evidence: A Live Demo of the Elicit API and MCP Server Presenter: Panda, Software Engineer at Elicit Date and Time: Wednesday, July 29, 2026, 10am PT Register: crowdcast.io/c/dxk71bqnsq2w

We rebuilt Elicit's search from the ground up and tested whether it actually was better. On BioASQ benchmark, Elicit had the highest recall at every result depth we tested from 10 to 200 results. Read the full write-up: elicit.com/blog/evalua...

Over the past year, Orion Pharma, a leader in human and veterinary medicines, has gone deep with Elicit. Serhii Vakal, PhD, Lead ML/AI Scientist at Orion Pharma, shared a compelling story about how he used Elicit. Read the full case study: elicit.com/customer-st...

In our latest fireside chat, Sheila Leunig, General Counsel at Elicit, and Zen Jelenje, Founder of Nascent Studio and former Data Solutions Lead at Elsevier, discussed how researchers can navigate copyright and data rights as AI becomes a larger part of research. Recording: youtu.be/6kQPwhjjhJ4

AI, Copyright, and Data Rights: A Framework for Scientific Research

AI tools like Elicit help scientists work faster and more accuratel...

youtube.com

The Elicit API and MCP server are now available and bring Elicit’s trusted capabilities to your agents and workflows. Three endpoints are now available: Search, Reports, and Systematic review. All three also work over MCP with Claude, ChatGPT, Gemini or any other MCP-compatible client. Demo below.

One day to go for the fireside chat on AI, Copyright, and Data Rights. The team will discuss topics such as data use, open access rights, other common licenses, and output ownership. Register today crowdcast.io/c/5b0xx25xfn16

At PEGS Boston, Xiangxu Kong, PhD, Scientific Leader at Elicit, shared how protein sciences teams are applying AI to published knowledge to support engineering decisions. www.youtube.com/watch?v=XEp...

AI-Powered Evidence Synthesis for Protein and Antibody Engineering

Antibody and protein engineering teams must navigate a rapidly expa...

youtube.com

Join Sheila Leunig, our General Counsel, and Zen Jelenje, Founder, Nascent Studio & former Data Solutions Lead, Elsevier, for a fireside chat covering data use and AI licensing in practice. AI, Copyright, and Data Rights July 14, 2026, 9:15 am PT Register: cca.st/5b0xx25xfn16

We just wrapped our fireside chat, "Beyond Benchmarks: Lessons from Building AI Evals in the Real World," with Pradyumna Prasad and Ben Rachbach. The recording is now live youtu.be/aPqyi45U1Ys.

Lessons from Building AI Evals in the Real World

As AI systems become more capable, evaluation is quickly becoming o...

youtube.com

Our Eval team's fireside chat is one day out. Beyond Benchmarks: Lessons from Building AI Evals in the Real World Tuesday, June 30, 10am PT (1pm ET) Save your spot: lnkd.in/gBsEprKr

Elicit is now in Anthropic's Connectors Directory. You can find it and connect in one click. Once connected, Claude can search 138M+ papers and run grounded literature reviews without leaving the chat. Elicit is also an app in ChatGPT. For Copilot, Gemini, and other MCP clients: docs.elicit.com.

Our Eval team is hosting a fireside chat on what evaluation looks like in the real world. Title: Beyond Benchmarks: Lessons from Building AI Evals in the Real World Date: Tuesday, June 30, 2026 Time: 10am PT (1pm ET) Save your spot today: crowdcast.io/c/xkrzz4885sak

Oncology target selection remains remarkably qualitative. At Bio-IT World 2026, our CEO Andreas Stuhlmüller shared how we're thinking about this problem. Watch the full talk: youtu.be/rTou2fsDwMc.

From Literature Search to Living Model: Causal Reasoning for Oncology Target Nomination

Despite billions in R&D investment, oncology target selection remai...

youtube.com

A fair question about any data extraction tool: can you actually trust what it pulls? Farhad Shokraneh, PhD, an SLR methodologist ran a scoping review for a policy brief and used Elicit to check the accuracy of each data point. Below he shares his experience with Elicit.

The recording is live for "What's Next in AI for Life Sciences: Evaluating Superhuman AI. AI is outpacing the evaluation playbook. The old metrics break down once AI takes on evidence synthesis, multi-step deliverables, and autonomous goals. Watch the recording here: youtu.be/ajUxiHmjORs...

What's Next in AI for Life Sciences: Evaluating Superhuman AI

AI in life sciences is moving faster than the evaluation playbook. ...

youtube.com

Monday's webinar is officially at capacity. If you didn't get a spot, fill out the form below and we'll send you the recording after the session. forms.gle/gqag7u4K42p... Jungwon will cover how to evaluate superhuman AI, why old metrics break down, and what a rigorous framework looks like.

We're heading to DIA from June 14-18 in Philly. Harry Johnson and Prem Mohanty from our team will be there for the week. Come find us at booth # 2026. We'd love to talk about how AI is actually changing your work.

Can Elicit's full-text screening recall every relevant paper in a systematic review? A sample of 74 Cochrane reviews and 377 studies was used for this evaluation. Elicit's results: Paper-level recall: 99.5% Per-criterion accuracy: 94.8%

As AI scales from basic tasks to complex automation, the ability to rigorously evaluate these tools is becoming the ultimate competitive advantage for life sciences teams. On Monday, June 15th, join Elicit Co-founder and COO Jungwon Byun for a live session. Register -> tinyurl.com/3ktyv94t

Does Elicit's abstract screening meet the sensitivity requirements of systematic reviews, while also maintaining high specificity? We evaluated Elicit on this stage using 108 Cochrane reviews. Our dataset included 931 studies that should be screened in, and 5,162 that should be screened out.

When we launched the upgraded Elicit systematic review last month, we talked a lot about speed. Under the hood, here are the key features doing the actual work: Search - Boolean and MeSH keyword queries alongside high-recall semantic search - Autotranslate between keyword and semantic search

How well does Elicit's semantic search work for systematic reviews? We sampled 888 Cochrane reviews across 12 MeSH areas. Elicit retrieved 95.0% of included studies on average. To learn how we evaluated AI search, read: elicit.com/blog/evalua... Try search on your own questions at Elicit.com.

In rare disease, every paper counts towards informing research questions. Patient cohorts are small, so a study from decades ago and one published last week can both shape current work. Heather Richbourg, PhD, at Ultragenyx, told us how she uses Elicit Systematic Review for this kind of work.

The full walkthrough of Elicit Systematic Review is now live. During the session, we ran a complete review from start to finish in 60 minutes while highlighting the cool new features that really made this feat possible. Try out your protocol at elicit.com. www.youtube.com/watch?v=MdM...

Protocol to PRISMA: A Live Systematic Review with Elicit

A walkthrough of completing an end-to-end systematic review in Elic...

youtube.com

Live tomorrow: a systematic review end-to-end in 60 minutes. We've been talking about what Elicit can do. Tomorrow we're going to show you. 11am–12pm PT Register: events.teams.microsoft.com/event/00a49...

Systematic reviews take 9 months on average. Can we do one live in 1 hour? Protocol to PRISMA: a Live Systematic Review with Elicit Wednesday, May 13, 11am–12pm PT Host: Hamsa Pillai, Product Manager Register → events.teams.microsoft.com/event/00a49...

The Elicit team (Xiangxu Kong and Prem Mohanty) is on the road at PEGS Boston 2026 next week (May 11-15). Be sure to attend the presentation by Xiangxu Kong to learn more about AI for literature search and evidence synthesis. Visit us at Booth # 35 for demos and discussions. See you there!