Frequently asked questions.
Everything you need to know about Kepler, our platform, and how we're building trustworthy AI, starting with finance.
What is Kepler?
Kepler is the ground truth platform for AI. It separates what AI does well (interpreting questions, understanding intent, generating responses) from what code does well (retrieving data, running calculations, citing sources). The result is AI that proves it's right: every number traces to its source, every calculation is reproducible, every conclusion reveals its reasoning.
How is Kepler different from ChatGPT or other AI tools?
General-purpose AI tools generate answers probabilistically. They can hallucinate numbers, fabricate sources, and give different answers to the same question. Kepler's architecture is fundamentally different: AI handles the language, deterministic code handles the data. Same input, same output, every time. Every figure links to its source document, page, and paragraph.
What is the semantic layer?
The semantic layer is the core of Kepler's architecture. It maps natural language concepts to precise, verified data definitions. When you say "revenue," "top line," or "total sales," they all resolve to the same canonical data point with a clear source. This means AI never guesses at what a number means or where it comes from.
What data sources does Kepler use?
Kepler Finance ingests data from SEC filings (10-Ks, 10-Qs, 8-Ks, proxy statements), earnings call transcripts, news feeds, web search, stock market data, and investor relations materials. We cover 950K+ SEC filings, 14K+ companies, and 40M+ IR documents across 27 markets.
What is traceability?
Traceability means every output links back to its source. Click on any number and see the exact filing, page, and paragraph it came from. Click on any formula and see how it was computed. Click on any conclusion and see the reasoning chain. This is Kepler's core differentiator: outputs you can defend, not just consume.
Who is Kepler built for?
Kepler Finance is built for buy-side analysts, private equity, investment banking, equity research, hedge funds, and compliance teams. Anyone who needs AI-generated financial analysis that they can verify and defend. We're starting with US public equities and expanding from there.
Can I use Kepler for non-US companies?
Kepler Finance currently focuses on US public equities. Support for international markets is on our roadmap. Contact us for details on timeline and coverage.
How does Kepler handle security?
Security is an architectural requirement, not a feature. Your data is stored in isolated environments, fully separated from other customers at every level of the stack. Every query, calculation, and data access is logged with complete lineage. We are GDPR and CCPA compliant, SOC 2 Type II certified, and pursuing ISO 27001 certification.
Does Kepler replace my existing tools?
Kepler complements your existing workflow. It's not a replacement for Bloomberg Terminal or FactSet as a data feed. Instead, it sits on top of verified data sources and gives you AI-powered analysis with full traceability. Think of it as the intelligence layer that makes your existing data actionable and defensible.
How do I get access?
Kepler Finance is live and in production. Request a demo through our website and our team will walk you through the platform and discuss how it fits your workflow.
Who built Kepler?
Kepler was founded by ex-Palantir engineers who built data infrastructure for the world's most demanding organizations. Our founding team includes senior engineers from Palantir, Citadel, and Meta. We're backed by the founders of Meta AI Research, OpenAI, dbt Labs, MotherDuck, and Square/Block.
Still have questions?
Reach out to our team or request a demo to see Kepler in action.