Elicit Blog

Introduction to pharmaceutical development by Senior Director of Global Health Economics and Outcomes Research at Pfizer

We believe three principles are especially important for making AI-powered research more collaborative

We evaluate Elicit search against web search engines on the BioASQ benchmark

Elicit Research Agent sets the frontier on our new decision-making benchmark for AI in the life sciences

Elicit Research Agent is an AI environment built for high-stakes decisions

Elicit outperformed five other search systems on BioASQ, retrieving more of the papers biomedical experts rely on.

The Elicit API and MCP server bring Elicit’s trusted capabilities to your agents and workflows.

Our Research Agent can do significantly more work and use more data sources, Elicit is moving from workflow-based limits to a monthly usage pool

Elicit Systematic Review now supports PRISMA 2020 guidelines, and is reproducible, traceable, and auditable at every step.

Elicit hit 95% search recall, 97% abstract screening, 99% full-text screening, and 96% extraction across 994 Cochrane reviews.

Models are getting better at long-horizon tasks. So why don't they help much when you're planning clinical development or thinking through a launch?

We've had more than 500 conversations with the top pharma companies and many external advisors. Here's what we believe about the future of R&D.

This is an excerpt from Elicit's April 2026 investor update. I take a step back, provide situational awareness, talk about hard-to-verify tasks, ...

Two years ago we published our stance on coding assistants in interviews. A lot has changed…

Search our 138M+ papers and generate Reports using our API.

AI elevates RWE planning into a corporate strategic mandate

The Arnold-Huxley debate on the role of sciences and humanities in education encapsulates some of the best arguments on both sides.

Claude Opus 4.5 is better than Sonnet 4.5, Google Gemini 3 Pro, and OpenAI GPT 5 at data extraction and writing reports with fewer hallucinations.

Elicit Research Agent powering workflows for competitive landscapes, research landscape and broad topic exploration

Let me start with a vision. Imagine you have a question about the future—say, "What would happen if California tried to secede from the US?"

Today’s practicing scientists and artists work at separate institutions and rarely collaborate. But has this always been the case?

Elicit achieves SOC 2 Type II certification.

Elicit Systematic Reviews now supports keyword search queries over Elicit, PubMed, and ClinicalTrials.gov.

Scientific literature is vast and contains within it as yet unnoticed connections. Literature-based discovery is an attempt to bring them to light.

Clinical Trials are now in Elicit. Search over 545,000 studies on clinicaltrials.gov and instantly summarize them into tables or reports.

Other automated alerts don't work: they're noisy and inefficient. Elicit Alerts leverages AI to surface relevant research and reduce wasted time.

Elicit's Systematic Review is our most advanced workflow, delivering human-level accuracy in evidence synthesis at a fraction of the cost and labor.

Elicit Reports provide automated research overviews with higher quality and efficiency than other deep research tools.

Elicit has raised $22M in Series A to expand beyond academia, building AI infrastructure for evidence-based reasoning across multiple industries.

Introducing Elicit Reports: fully automated rigorous research overviews, designed for researchers, and built on the principles of systematic reviews.

Elicit Systematic Review turns months of manual work into minutes, guiding researchers through each step while preserving rigor and accuracy.

Reinforcement learning may boost LLMs today, but it cannot deliver safe, long-term intelligence. We argue for System 2 learning instead.

Elicit develops LLM-based auto-evals to balance scale, trust, and flexibility, ensuring reliable scientific reasoning at superhuman speed.

Beyond ladders: a polynomial model of career growth.

Not every AI role needs a PhD in ML. A good AI engineer can bridge full-stack skills with LLM expertise, making AI practical in real products.

Elicit implements SPLADE to leverage semantic search while retaining consistent rigor and accuracy.

Elicit is rethinking UX with living documents: complex, editable objects that leverage AI to turn our scientific knowledge into an evolving record.

Elicit introduces Notebooks: a new way to analyze research by unifying search, summaries, and systematic review on a single, living page.

Elicit has launched a new user-facing feature nearly every week since 2021, making iteration at speed our core practice. We know how to do it well.

An argument for "alignment windfalls": when safer AI systems don’t just reduce risk, but generate immediate value for companies and society.

Elicit's nuanced stance on coding assistants in interviews: welcome as tools for testing insight, but not as replacements for genuine reasoning.

Don’t confuse embeddings with memory. To build effective RAG, start with a search engine, not just a vector database.

We evaluate an automated approach for catching hallucinations in paper abstracts, aiming for consistently more trustworthy results.

Elicit’s two main contributions to AI Safety are strengthening epistemics and prioneering advanced process supervision.

Many AI assistants promise ease of use. Elicit distinguishes itself by making reasoning systematic, transparent, and unbounded.

Our mission at Elicit is to scale up good reasoning — because better reasoning leads to better decisions, at every level.

With $9M in new funding, Elicit advances its mission: using language models to make scientific reasoning faster and more reliable.