Ethos vs ValidMind (2026)
Ethos and ValidMind both sell to bank model risk teams, but they start from different ends. ValidMind is a validation platform. Its open source library runs tests inside the bank's environment, its GPT-4.1 features draft documentation for a validator to check, and it plugs into SageMaker, Databricks, Snowflake and MLflow, where models are built. Ethos is a workflow layer across six risk functions, and model risk is one of them. Its agents gather evidence and draft assessments on the AI providers a bank has already approved, and validators make the calls. ValidMind carries named proof, with Experian selling an assistant built on it and General Bank of Canada reporting a 70% cut in validation time. Ethos names no customer and no integration. Pick ValidMind for a model inventory tied to the tools data scientists use, with templates mapped to SR 26-2, SS1/23 and OSFI E-23. Pick Ethos to run model risk, control testing and call center review on one platform, self hosted if needed, with agents on models the bank already trusts.
- You want model risk, control testing and call center review on one platform. Ethos runs six solutions on a shared Data Layer and Workflow Engine, from generative AI governance to change reconciliation.
- You want agents on models you have already approved. Ethos agents run on the AI providers and in house models a bank has connected, rather than on a model Ethos picks.
- Your data has to stay in house. Ethos offers a self hosted installation or a dedicated private cloud alongside its managed cloud, and the client keeps control of its data.
- You want to widen autonomy step by step. Each agent action can be set to run automatically, need approval or be blocked, and every action by a person or an agent lands in one audit trail.
- You want validation effort sized to risk. Ethos tracks every model, vendor model and AI application with its tier, materiality and owner, and sends high materiality models to full independent validation.
- Your models are built in SageMaker, Databricks or Snowflake. ValidMind connects to those platforms and to MLflow, and raises issues in Jira and ServiceNow, so the inventory follows the work.
- You want tests run where the data lives. ValidMind's open source Python library runs inside your environment and sends results, not datasets, to the platform.
- You answer to supervisors in several countries. Its templates map to SR 26-2, the PRA's SS1/23, OSFI E-23 and the EU AI Act, with Japanese and insurance guidance also covered.
- You want proof from named customers. General Bank of Canada's chief risk officer reports a 70% cut in validation time, and Experian sells an assistant for model risk management built on ValidMind.
- You are putting AI agents into production. Agent Authority, in early access, sets each agent action to run alone, ask first or never run, and scopes every approval to one action.
This comparison is published by AI FinTech Index, an independent research platform that publishes independent ratings of AI vendors for financial services. Ethos and ValidMind are each graded against the same capability taxonomy, from each vendor's own public materials and the regulatory record, under the AI FinTech Index verification standard. No vendor pays for placement, and no vendor has reviewed this page. How this evidence is graded
Plain facts
| Ethos | ValidMind | |
|---|---|---|
| Primary category | Compliance, Surveillance & RegTech | Compliance, Surveillance & RegTech |
| Founded | 2023 | 2022 |
| Headquarters | New York, New York, United States | Palo Alto, California, United States |
| Website | ethosai.com | validmind.com |
Side by Side
Select any grade to read the note behind it.
| Axis | E Ethos |
V ValidMind |
|---|---|---|
| AI Centrality | ||
| Autonomy and Oversight Model | ||
| Model Risk Management and Transparency | ||
| Operational and Outcome Evidence | ||
| AI Safety and Data Stewardship | ||
| GLBA and Data Privacy Posture | ||
| Security Certifications and Trust Center | ||
| Regulatory Status and Licensure | ||
| AI Governance and Bias Disclosure | ||
| AI Liability and Recourse | ||
| Model Supply Chain Disclosure | ||
| Core Systems and Integration Depth | ||
| Deployment Model and Data Residency | ||
| Commercial Transparency | ||
| Institution and Segment Coverage |
The short version of each
Ethos
Ethos AI is a New York company that sells agent governance and workflow software to banks. Model risk is one of six solutions on its platform, alongside control testing, call center review and generative AI governance. Its agents gather evidence, draft assessments and route approvals on the AI providers and in house models a bank has already approved, while validators and testers make the decisions. Banks can run it self hosted, in a dedicated private cloud or as a managed cloud. Canapi Ventures led its $6 million seed round, with Capital One Ventures and Better Tomorrow Ventures. The caveats are proof and security. No customer, integration or result is published, and no security attestation or trust center is available.
Source: AI FinTech Index, 2026
ValidMind
ValidMind, a Palo Alto company launched in 2022, sells model risk management and AI governance software to banks and insurers. Its open source Python library runs model tests inside the customer's environment, and features built on OpenAI's GPT-4.1 draft documentation and evidence assessments for a validator to review. It connects to SageMaker, Databricks, Snowflake and MLflow, and maps its templates to SR 26-2, SS1/23, OSFI E-23 and the EU AI Act. Experian sells an assistant for model risk management built on ValidMind, and General Bank of Canada's chief risk officer reports a 70% cut in validation time. The caveats are price and security. ValidMind publishes no price. The AI FinTech Index explains that its SOC 2 claim comes with no report type or audit.
Source: AI FinTech Index, 2026
Common questions
How does Ethos compare to ValidMind?
Both sell to bank model risk teams. ValidMind is a validation platform with an open source test library, drafting features built on GPT-4.1 and integrations with SageMaker, Databricks and Snowflake. Ethos is a workflow layer across six bank risk functions, with agents that gather evidence and draft assessments on the AI providers a bank already uses. ValidMind names customers and partners, including Experian and General Bank of Canada, while Ethos names none. Graded by AI FinTech Index against the same capability axes from each vendor's own published materials, verified November 3, 2026. No vendor pays for placement.
How much do Ethos and ValidMind cost?
Neither publishes a price. ValidMind sells four packages, from AI governance to development automation, priced mainly on how many models and AI systems a bank governs, and it is also sold through a private AWS Marketplace listing. Ethos publishes no price, tier or billing unit and offers a demo request, so a buyer cannot tell whether billing would follow models, solutions or users. Graded by AI FinTech Index against the same capability axes from each vendor's own published materials, verified November 3, 2026. No vendor pays for placement.
Which model risk guidance do Ethos and ValidMind support?
Both align with SR 26-2, the interagency model risk guidance issued in April 2026 that replaced SR 11-7. Ethos also names OCC Bulletin 2026-13, the federal stress testing programs and the NIST AI risk framework. ValidMind adds the PRA's SS1/23, OSFI E-23, the EU AI Act, NAIC guidance for insurers and the Japanese FSA's principles. Neither Ethos nor ValidMind claims a license or supervisory status of its own. Graded by AI FinTech Index against the same capability axes from each vendor's own published materials, verified November 3, 2026. No vendor pays for placement.
Where do Ethos and ValidMind run, and what happens to bank data?
Ethos offers a self hosted installation, a dedicated private cloud or a managed cloud, and the client keeps control of its data. ValidMind runs as a shared or single tenant cloud on AWS, Azure or Google Cloud, with the region set in the contract, and its test library keeps datasets inside the bank's environment. Neither Ethos nor ValidMind publishes a subprocessor list, and neither says whether one bank's findings shape what another bank receives. Graded by AI FinTech Index against the same capability axes from each vendor's own published materials, verified November 3, 2026. No vendor pays for placement.
Who makes the call when Ethos or ValidMind uses AI?
In both, the AI drafts and a person decides. Ethos agents gather evidence and draft assessments while validators and testers make the calls, and each agent action can be set to run, need approval or be blocked. ValidMind presents its drafts as a starting point for validator review, with approval steps and digital sign off in its workflows. Neither Ethos nor ValidMind publishes an accuracy figure for its drafts or a threshold for when one can pass without review. Graded by AI FinTech Index against the same capability axes from each vendor's own published materials, verified November 3, 2026. No vendor pays for placement.
Related comparisons
Other published head to head assessments involving these vendors or their closest peers. The full set for this category is on the Credit Decisioning & Underwriting page.
Neither vendor publishes a price, a subprocessor list or an audited security report. ValidMind says its controls are designed to align with SOC 2 but names no report type, and Ethos publishes no certification at all. For Ethos, the open questions are which model provider and hosting serve its own drafting, which banks run it today and which systems it connects to.
For ValidMind, they are its SOC 2 report, its list of hosting regions, and whether GPT-4.1 is reached through OpenAI or Azure OpenAI. Neither says whether one bank's findings shape what another bank receives, how often its drafts need correcting before sign off, or what fairness testing covers its own AI.