Frequently Asked Questions (FAQ)

Kineviz Agent is an AI agent for data analysis built for people who can’t take AI at its word. Every observation it makes stays connected to the source data behind it, so analysts can check the work themselves. It performs conversational graph analysis and builds custom applications — dashboards, workflows, and reports — inside Kineviz. The Community edition is free, with a Pro upgrade for power users.

The Kineviz approach to AI is "human-centered intelligence". The platform is designed so that AI agents and visual analytics work together to surface relationships and insights. We believe that human questions should shape data models and AI tools should augment, but not replace human judgment.

Kineviz Agent treats each observation as a claim about your data, rather than a statement of fact, and delivers it with the supporting records — select a finding and see exactly which nodes, edges, and fields it drew from. Analysts can review the evidence, not just the summary, and can confirm, challenge, or throw out any finding.

The most reliable safeguard is source traceability — requiring every AI observation to link back to real records. Kineviz Agent is designed around this: every observation stays connected to the original data behind it, every step it takes is recorded, and every conclusion is traceable. Instead of trusting a summary, analysts inspect the source data each finding drew from, and every conclusion presents its own evidence.

An AI agent can run analyses you describe in plain language. With Kineviz Agent, you ask for what you want to see — for example, "show me every account within two hops of the flagged wallet, sized by transaction volume and colored by transaction type" — and it runs the analysis, builds the model, and draws the visualization in your graph, handling filtering, community detection, and time-series charts.

Yes. Kineviz Agent builds applications inside Kineviz through plain-language requests — custom interfaces, dashboards, workflows, and reports shaped to your team’s investigation. For example, you can ask it to "build a triage dashboard for new alerts, with a one-click case report" and get a ready-to-use tool for your analysts.

With Kineviz Agent, yes — the agent accelerates the work while analysts keep final judgment. It rapidly finds connections, traces paths, and surfacesg anomalies, but every step is visible and every finding comes with its supporting evidence. Analysts confirm, challenge, or discard each observation with the source data in front of them as required by investigative standards.

Both. Kineviz Online (formerly GraphXR) runs in the browser, while Kineviz Desktop brings the same workspace to your computer with an embedded graph database and a built-in AI agent. Teams can launch Kineviz Online directly from kineviz.com or download Kineviz Desktop, choosing the deployment that best fits their data and security requirements.

Kineviz Desktop is a workspace for connected data analysis that runs on your own computer. It is available as a free download from kineviz.com. It ships with an embedded (KoreDB) graph database and includes Kineviz Agent, an AI agent built for investigative work, and provides in-app sandboxes and tutorials covering first steps to advanced analytics. Drop in a spreadsheet or folder and it becomes a graph in minutes. If you already run BigQuery Graph, Neo4j, or another graph source, Kineviz Desktop connects directly so you can analyze the data in place.

Yes — your data and your analysis workspace stay on your computer. Kineviz Desktop includes Kineviz Agent, an AI agent built for investigative work on connected data. Ask it to find connections, trace paths, or surface what’s unusual in your graph. You stay in control: every step it takes is visible, and nothing goes into a report that you can’t verify. The AI model itself runs as a cloud service. Organizations that need the model on their own infrastructure can host it locally with Kineviz Enterprise.

Yes. If you’re already running Google BigQuery Graph, Spanner Graph, PuppyGraph, UltipaGraph, Neo4j, or another graph or big data store, both Kineviz Online and Desktop connect directly and let you analyze the data in place — no export or migration needed. For data that isn’t in a graph database yet, Kineviz turns relational data, spreadsheets, CSVs, and folders of files into graphs in minutes. Ask the agent, or model using visual transforms. The embedded KoreDB database keeps the result local.

The hardware and software requirements are for clients.

An enterprise graph analytics platform lets analyst teams explore connected data under security controls that large organizations require. Kineviz Enterprise is graph analysis deployed on your terms: self-hosted, with SSO, role-based and row-level access controls, and the audit logging needed by investigation teams in financial services, forensics, and government.

Yes. Kineviz Enterprise runs inside your own infrastructure, on-prem or in your private cloud, so your data stays behind your firewall, under your policies, on hardware you control. The self-hosted deployment model is built for organizations whose data can’t leave the building, including regulated industries and government agencies.

Yes. Kineviz Enterprise supports fully air-gapped deployment - no outside connectivity at all. It is designed for regulated, classified, and sovereign environments: you choose where your data lives and where it’s processed — a specific region, a specific country, or an isolated network. Agencies and regulated industries with residency requirements can run the same graph analysis capabilities inside boundaries they define.

Kineviz Enterprise deploys within the boundaries your organization defines, giving you control over where data lives and where it’s processed. Options range from deployment in a specific region or country to a fully air-gapped network with no external connectivity. Combined with self-hosting behind your firewall, this addresses residency and sovereignty requirements for government agencies and regulated industries without giving up graph analysis capability.

Investigation platforms need role-based and row-level controls plus a full audit trail, and Kineviz Enterprise provides all three. You can define roles with specific privileges — admin, analyst, read-only reviewer, or roles shaped to your workflow — and restrict visibility down to individual rows, so two analysts in the same project each see only records they’re authorized to see. Audit logging records who accessed what, when, and what they did.

Yes. Kineviz connects to your identity provider over SAML or OIDC, so your existing authentication policies — MFA, session rules, and instant offboarding — apply automatically. Administrators also get a central console for managing users, permissions, deployments, and policies, giving your organization the visibility and levers to run Kineviz as your security team requires.

Kineviz keeps enterprise pricing simple, and you’ll hear it on the first call. There’s no drawn-out discovery process before you get a number. If your evaluation involves a security questionnaire, the Enterprise offering covers self-hosted and air-gapped deployment, SSO over SAML/OIDC, role-based and row-level access controls, audit logging, and admin controls, so procurement and security review can move quickly.

Kineviz Explorer is an offering through the Google Cloud Platform marketplace to connect Kineviz either to a Google Big Query data warehouse or Google Spanner relational database.

Link analysis software helps law enforcement by turning information residing in both structured and unstructured documents into a connected view of how suspects, assets, and events relate, exposing networks that isolated records hide. With Kineviz, investigators retrieve and structure digital footprints, social and professional connections, commercial interests, and assets into a navigable reasoning graph—accelerating investigative decision-making with improved confidence.

OSINT investigators use graph visualization platforms to map relationships between people, accounts, and events gleaned from open source information. As such, Kineviz equips analysts to rapidly gather, fuse, and interpret both OSINT and proprietary data — its SeekerXR offering reaches more than 80 OSINT and proprietary sources — combining closed-web data, open sources, and AI-enhanced document analysis into a unified graph of relationships investigators can explore visually

Graph visualization maps relationships between threat signals so security analysts can see hidden networks and emerging risks, instead of scanning isolated alerts. Kineviz turns fragmented signals into a connected intelligence layer AI can reason over, letting cybersecurity teams explore how entities connect and surface emerging threats with evidence-backed insight.

Yes— a published Kineviz case study describes using network visualization to deplatform an extremist group from social media, demonstrating how mapping relationships across accounts supports concrete enforcement outcomes.

Yes—location and timing are often the connective tissue of a case, and folding them into an analysis of relationships reveals patterns neither shows alone. A published Kineviez case study shows how investigators interpreted where and when events occurring within a connected graph.

Investigators combine OSINT and proprietary data by fusing them into a single relationship graph where entities from every source connect. Kineviz unifies closed-web data, OSINT sources (SeekerXR alone covers 80+ OSINT and proprietary sources), and AI-enhanced document analysis into one graph; AI handles extracting and connecting signals across sources, while investigators visually trace relationships, validate findings, and refine hypotheses in real time.

A defensible investigation requires every conclusion to be traceable to evidence. To achieve this, Kineviz keeps humans in the loop: AI extracts and connects signals and their underlying sources and presents the results as a network graph. Investigators then visually trace each relationship, validate findings against sources, and refine hypotheses themselves. The graph enables faster threat detection along with transparent, defensible, and auditable investigations.

Graph analytics spotlights fraud by revealing relationship patterns such as shared accounts, coordinated behavior, and money flows that row-and-column BI tools miss. With Kineviz this is a visualization-first approach: analysts explore relationships in multi-dimensional graph views, use interactive filtering to remove noise, and trace AI-derived insights back to source data. When a case needs outside context, Kineviz SeekerXR brings in open-source intelligence from 80+ OSINT and proprietary sources.

Graph databases store data as entities and relationships, so multi-hop connections that reveal fraud rings stay intact instead of being flattened into rows and joins. Kineviz provides visualization and analytics in a navigable system that both AI and human analysts can reason over, making it faster to uncover hidden patterns and validate findings. When an online gaming platform needed to find fraudulent activity lurking in its stream of 40 million events per day, standard BI tools were overwhelmed. Using Kineviz with real-time big data analytics, evolving collusion and fraud threats were detected early enough to act before losses mounted.

Yes — and the right role for AI is amplifying investigators, not replacing them: it accelerates fraud detection, but trustworthy results require transparency, not black-box scores. The Kineviz AI Agent is built for law enforcement, fraud investigation, intelligence, and compliance, where findings must meet an evidentiary standard. The Agent can quickly build a network graph that augments the analyst’s field of view, recording every step it takes and keeping a complete trail from each conclusion back to the supporting source data. With this information, an analyst can highlight results that are explainable, auditable, and backed by real evidence.

A graph-based fraud workflow is iterative: connect data, visualize relationships, filter noise, test hypotheses, and document evidence. With Kineviz , analysts combine intuitive interfaces with multi-dimensional graph visualization and interactive filtering, accelerating iteration from initial signals to decision-ready findings.

No. Kineviz includes no-code, point-and-click Cypher queries. Combined with seamless data integration and visual exploration, fraud teams can explore relationships directly, without needing to translate every investigative question into code.

Graph-powered business intelligence analyzes data as a network of connected entities rather than static charts and tables, revealing the relationships behind the metrics. Kineviz presents your data as a connected graph that AI can reason over, so you see not just static metrics, but how entities, signals, and dependencies relate—​surfacing opportunities and emerging risks.

Traditional BI aggregates data into dashboards that show what happened; graph analytics preserves the relationships between data points so you can explore why. Kineviz describes it as dashboard meets mind map: structured metrics combined with a connected data model you can navigate at the forest, tree, and leaf level, instead of static side-by-side charts.

Traditional BI only takes you as far as correlation; identifying cause and effect requires analyzing relationships between entities. Kineviz addresses this with AI-assisted, graph-powered analysis. By preserving structure, it reveals clusters, paths, and causal patterns that disappear in isolated charts, helping analysts drill down from correlated metrics to the mechanisms behind them.

You extract value from a data lake by making its contents easy to connect, explore, and visualize. Kineviz helps reclaim the data lake by turning existing data into a connected, explorable graph, with visual transforms for cleaning and modeling data on the fly, so analysts can test hypotheses quickly without lengthy data engineering cycles.

AI for BI is making analysts dramatically more effective by moving analysis from static reporting to relationship-driven reasoning: finding patterns, testing hypotheses, and explaining outcomes. Kineviz turns charts and tables into a connected graph AI can reason over, within a shared, explainable reasoning environment where analysts can explore context, validate assumptions, and refine insights.

Hidden risks and black swans usually emerge from interdependencies between entities, which static dashboards tend to flatten away. Kineviz helps analysts spotlight risks by connecting entities, signals, and dependencies into a navigable graph. This lets teams monitor interdependencies across many levels of detail and detect emerging risk patterns in the relationships behind their metrics before they surface in headline numbers.

Yes. Kineviz customers in life sciences and healthcare enterprise include Thermo Fisher, Cepheid, Octave Bioscience, Rally Health, and the National Institute of Allergy and Infectious Diseases and National Institutes of Health. These organizations use Kineviz’s visual graph analytics to connect and explore complex biomedical and research information.

A healthcare knowledge graph connects entities like diseases, treatments, markers, and publications into a network of relationships that researchers and AI can reason over. Kineviz can build unified live knowledge graphs that extract and connect the key information available in these sources.

Knowledge graphs accelerate biomedical discovery by revealing relationships across literature, experiments, and observations which would remain hidden in isolated sources of information. With Kineviz, researchers can track novel markers in a multi-dimensional graph view that surfaces emerging patterns, trace each marker back to its source publications. This provides research teams with transparent, relationship-driven exploration and analysis in a collaborative exploratory workspace.

Patient journey mapping visualizes a patient’s experiences and clinical events over time so teams can understand progression, context, and outcomes. Kineviz supports mapping of patient journeys, which can include complex data from clinical trials, advanced 3D imaging, and other healthcare innovations.

Graph analysis links imaging findings to patient context and derived features, so researchers can study how changes relate over time instead of just comparing scans. Kineviz can expose the temporal and topological attributes of progressive diseases by connecting MRI findings into a navigable graph, helping researchers resolve ambiguities and communicate evidence-backed insights to clinicians and patients.

AI automates the extraction of entities and relationships from peer-reviewed publications, turning unstructured text into a structured, connected information. In Kineviz, AI-extracted relationships join experimental data and observations in a unified knowledge graph that research teams can navigate collaboratively. Patterns hidden in isolated tables are revealed while keeping every finding traceable to its source.

Graph visualization helps clinical teams see relationships across trial data relating to patients, timestamps, findings, and derived features. When applied to complex clinical trial and imaging data, this can expose temporal patterns in disease progression and make results easier to explore, validate, and communicate clearly across teams, clinicians, and patients.

Effective collaboration requires a shared, navigable structure where information links back to its evidence. In Kineviz this is presented as a collaborative knowledge graph: researchers explore connected literature, experimental data, and observations together, trace novel markers to their sources, and communicate evidence-based insights clearly across teams.