A Question From the Future
Picture the scene, three years from now. A tax auditor sits across the table, points at the interest rate in your intercompany loan documentation, and asks: “Where does this number come from?”
In 2026, the honest answer at more and more firms includes the words “an AI prepared the analysis.” That answer is fine. Auditors assess documentation on its content, not on who typed it. The follow-up question is the one that decides the audit: can you show the path from raw market data to that rate, step by step?
If yes, the AI saved you weeks. If no, you are defending text that sounds right and proves nothing.
AI Has Arrived in Transfer Pricing
The shift is real and it is fast. AI agents today read loan agreements, screen comparables, run rating analyses, and draft benchmarking studies. Transfer pricing was made for this: the judgment sits in a few key decisions, while the hours sit in gathering data, running screens, and writing everything up. Agents take over the hours. Professionals keep the decisions.
The tools are here too. Supernomial, an AI-native platform for transfer pricing teams, lets professionals direct agent teams that prepare local files and financial transactions studies end to end.
Which brings the data question to center stage. An agent works at the speed of software, and it will happily work at that speed on bad data. The quality gate has to sit underneath the AI, in the database itself.
Accountable Data: Our Answers
We call our standard accountable data, and it rests on three commitments we have engineered into the platform since 2017:
- Primary sources only. Price data straight from global exchanges, financials from audited statements and registries, interest rates from central banks. More than 30,000 listed companies across 100+ capital markets, over 100,000 active and historical corporate bonds, day-precise back to 2010.
- A calculation log on every output. From beta factor to credit spread, each result documents its full path from raw data to number. Your auditor can walk the same path. No black box.
- Independence. smartZebra holds no advisory mandates, so no benchmarking result serves a second agenda.
For a decade, auditors, tax advisors, and CFOs have used this foundation to source margin benchmarks, credit spreads, and arm’s-length interest rates. The AI era does not retire that foundation. It multiplies its value, because now the data feeds agents as well as analysts, and the traceability travels with every study the agents produce.
Proof in Practice: smartZebra x Supernomial Partnership
The clearest signal that accountable data and AI belong together: Supernomial chose smartZebra as data partner for its agentic transfer pricing workflows.
In the first joint solution, Supernomial’s AI agents prepare complete financial transactions studies on smartZebra data. Take an intercompany loan: the agents assess the borrower’s debt capacity, derive a credit rating from quantitative and qualitative factors, apply notching for implicit group support, screen the bond universe by sector, country, and currency, and document an arm’s-length interest rate. One workflow, end to end, and each figure inherits the smartZebra calculation log.
Your team defines the playbook once, reuses it for each loan, and spends its time where it belongs: challenging inputs and signing off on ranges.
About smartZebra
smartZebra GmbH is the database for business valuation and transfer pricing. Founded in 2017 and based in Germany, smartZebra consolidates every capital market parameter that valuation and transfer pricing work requires: beta factors, valuation multiples, risk-free rates, credit spreads, and margin benchmarks. The data covers more than 50,000 listed companies across 100+ capital markets and over 100,000 corporate bonds, with day-precise history back to 2010 and daily updates. Every output ships with a calculation log that traces the result to its raw data, and exports are audit-ready (XLSX, CSV, PDF). Audit and tax firms, corporates such as E.ON and Deutsche Bahn, fund administrators, and tax authorities work with smartZebra under IDW S 1, IFRS, and the OECD Transfer Pricing Guidelines. A free 5-day trial with full platform access is available at http://www.smart-zebra.com.
About Supernomial
Supernomial provides managed AI agents for transfer pricing consultancies and in-house tax teams. Professionals direct their agent teams inside whichever enterprise AI system they already use, from Claude Cowork to ChatGPT Work to Microsoft Copilot Cowork. The agents prepare transfer pricing deliverables, from local files to financialtransaction studies. Every agent team runs on the customer's own methods and institutional knowledge, and that knowledge stays theirs. Learn more at http://www.supernomial.co.
Related pages
Last updated: August 4, 2026
