How To Use SciSpace and Copilot - Dominate Research in ONE tool!

SciSpace is positioned as an all-in-one research workspace that turns literature review workflows—finding papers, extracting key claims, and drafting notes—into a single semantic-search and AI-assisted loop. The core value centers on literature review: instead of keyword-only searching, users type a research question and get AI-generated summaries grounded in published papers, with options to filter by access type (PDF/Open Access) and by journal/conference tier. A standout feature is the “Insight from top five papers,” which provides a quick take-home message while still letting users click through to the underlying references.

The literature review interface is built around a results table that makes provenance visible. Multiple columns can be added to generate structured outputs—such as conclusions, abstracts, and limitations—across the set of retrieved papers. The workflow is designed for triage: users can start with a short “too long didn’t read” summary to decide whether a paper is worth deeper attention, then drill into specific sections. When AI-generated fields take time, users can leave the page and return later as the missing columns populate, keeping the process moving rather than forcing a wait.

Beyond discovery, SciSpace’s “My Library” supports question answering across documents users upload. This is aimed at multi-paper synthesis—something researchers often want when they have several PDFs and need answers that draw from all of them, not just one document. The transcript emphasizes that extracted data becomes the backbone for later features: after uploading a PDF, users choose which columns to extract (and can add custom columns), then use the resulting summaries as the structured dataset for further AI-assisted research.

SciSpace’s PDF “Copilot” adds another layer: users can ask questions directly about a specific paper, generate summaries of targeted sections (like the introduction), and get responses with citations that link back to where the information appears in the document. The interface also supports practical reading aids—highlighting text, explaining math and tables by interpreting images of equations/figures, and suggesting related papers. While “high quality” responses are described as behind a paywall, the standard responses are presented as usable for many tasks.

A major expansion is “Notebook,” described as a new workflow hub that can store and transform outputs from Copilot. Users can highlight text and send it to an “AI writer” for actions like simplifying, summarizing, translating, fixing grammar, or generating opposing arguments, then build sections and paragraphs from those drafts. Notebook entries can also be created by saving Copilot responses, effectively turning extracted bullet points into a literature-review outline. The transcript also highlights integrations—especially importing from Zotero to sync libraries into SciSpace—and a pricing tier that includes unlimited Copilot messages, customized Copilot answers, and the library features for $12/month when billed annually. Overall, SciSpace is framed as a research operating system that compresses search, extraction, synthesis, and drafting into one place, with Notebook positioned as the bridge from paper-level insights to publishable structure.