# Holistic AI

> Source: https://agentsindex.ai/holistic-ai

Enterprise AI governance that discovers shadow AI, red teams it, and has Guardian Agents block unsafe actions live.

Holistic AI is an enterprise AI governance platform that discovers models, agents and applications across cloud, code and SaaS, then tests them for bias, security and robustness. Guardian Agents watch running systems and intervene when a policy threshold is crossed.

- Best for: Enterprise governance teams who need continuous visibility and control over AI systems.
- Look elsewhere if:
  - you want to use it for real without paying
  - you want to start without talking to sales
  - you need the full API on every plan

Starting price

On request

From [holisticai.com](https://www.holisticai.com/), checked 28 October 2026

- MCP server Not stated
- API Partly From holisticai.com/enforce, 28 October 2026 [holisticai.com](https://www.holisticai.com/enforce)
- Open source Yes Checked by us github.com/holistic-ai/holisticai/blob/main/LICENSE, 15 November 2026 [Checked by us](https://github.com/holistic-ai/holisticai/blob/main/LICENSE)
- Self-host Not stated

What we checked

- Pages last read: 15 November 2026
- Facts checked: 16 facts · From 8 sources
- AI answers: Too few read · 58 buyer-question answers read, November 2026

[How we check](https://agentsindex.ai/methodology) [Editorial policy](https://agentsindex.ai/editorial-policy#not-for-sale)

[Visit Holistic AI](https://agentsindex.ai/go/holistic-ai) [Compare](https://agentsindex.ai/compare?cat=security-ai-agents&products=holistic-ai)

## At a glance

- Category: [Securing AI use](https://agentsindex.ai/categories/security-ai-agents)
- Tags: EU AI Act, GDPR, ISO 27001, ISO 42001, Private AI, SOC 2

### For developers

- Works with: AWS, Azure, GitHub, Databricks, Google Cloud, LangSmith and 20 more

## Holistic AI in our research

[All research](https://agentsindex.ai/blog)

Study 19 October 2026 – 16 November 2026

### [AI Tool Pricing Index, November 2026](https://agentsindex.ai/data/ai-tool-pricing-index)

Pricing

Contact sales

82 of the 361 tools we checked ask you to contact sales for a price.

Edition price, checked 28 October 2026.

Pricing pages of the AI tools, November 2026 edition as of 16 November 2026 n = 361 tools checked

## Who's behind it and how long it's been running

Holistic AI is built by Holistic AI Limited, a UK company (Companies House #12962678) incorporated on 20 October 2020 and registered in London. It was founded by Emre Kazim and Adriano Koshiyama, who met while working on AI ethics in UCL's computer science department before spinning the company out. Kazim has published more than 50 peer-reviewed articles on AI governance, and the company still runs an active research arm (HAI Lab) that has published around 80 papers on bias measurement, LLM auditing and agentic-system testing — an unusual amount of public research output for a governance SaaS vendor, and it's the methodology base the commercial testing features are built on. Company funding and financial scale are not established from the sources we could check today.

## Independent recognition: a real Gartner placement, not just a badge

Gartner published its first Magic Quadrant for AI Governance Platforms in 2026, evaluating 13 vendors; Holistic AI was the only one named a Challenger. In the companion Critical Capabilities report, Holistic AI ranked #1 for the AI Risk and Compliance use case and placed in the top three for AI Agent Governance. This comes from Holistic AI's own blog post, which links to the official Gartner reprint pages rather than just asserting the ranking — a genuine third-party evaluation, though we have not read the full Gartner report ourselves and are relying on the vendor's summary of it.

## Security and compliance posture — what's actually certified

Holistic AI publishes a detailed public Trust Center, not just a badge wall. As of today it holds ISO/IEC 27001:2022 certification (92 of 93 Annex A controls in scope, zero nonconformities at its 2026 certification audit) and a SOC 2 Type 2 report covering the Security, Availability, Confidentiality and Privacy criteria, both released to prospects under NDA. It is UK-GDPR compliant, acting as data controller for its own site and data processor under customer contracts. One thing worth flagging given this listing carries an iso-42001 tag: Holistic AI itself does not hold ISO/IEC 42001 certification — its own Trust Center lists that one as 'in progress,' not certified. The platform can map a customer's own controls to the ISO 42001 standard; that's a product feature, not a claim that the vendor is certified against it.

## What the platform actually covers

The product is organized around three modules — Identify, Protect, Enforce — plus a supervisory 'Guardian Agents' layer: Sentinel Agents that watch AI behavior for prompt injection, jailbreaks, data leakage and bias, and Operative Agents that can block or redirect actions in real time when a threshold is crossed. Identify connects, read-only by the vendor's description, to cloud platforms (AWS, Azure, Google Cloud), code repositories (GitHub, GitLab, Bitbucket), ML tooling (Databricks, MLflow, Weights & Biases), LLM providers (OpenAI, Anthropic, Google AI) and agent frameworks (LangGraph, CrewAI, AutoGen) to build a live inventory, including shadow AI. Protect runs automated testing (bias, robustness, privacy, transparency) and agentic red teaming. Enforce maps controls to the EU AI Act, NIST AI RMF, ISO 42001 and NYC Local Law 144, turning them into approvals, audit trails and deployment gates.

## Customers are verifiable, not just a logo wall

Unlike many governance vendors that show customer logos with no further detail, Holistic AI publishes named case studies with specifics: Starling Bank commissioned a third-party assessment of its customer-onboarding AI; the Wikimedia Foundation used Holistic AI to audit Wikipedia under the EU Digital Services Act; Allegis Group and MindBridge used it for enterprise AI trust and third-party AI auditing; recruiting-tech vendors Bryq and Hired used it for New York City Local Law 144 bias audits. Unilever, Mapfre, Johnson Controls, GSK, Siemens, eBay, Aon and several others appear as logos on the site without a published case study, so we could confirm they're customers but not independently verify how deeply each one uses the platform.

## Pricing and how to evaluate it before a sales call

Holistic AI does not publish pricing; there is no public pricing page, and every path on the site leads to a demo request. It also does not publish a public API reference or developer docs — we checked its full sitemap for a docs or api subdomain and found none. For a buyer comparing governance platforms, this means the kind of technical evaluation that's normally self-serve elsewhere — what a test actually checks, how a red-team report is structured, how the connectors authenticate — is only available once you're already in a sales conversation.

## EU AI Act timing, since that's one of this listing's tags

Independent of the vendor: the EU's Digital Omnibus, in force since 27 July 2026, pushed back the AI Act's high-risk obligations — standalone Annex III systems (recruitment, credit scoring, education, law enforcement) now have until 2 December 2027, and AI embedded in already-regulated products (Annex I) until August 2028. The Article 50 transparency duties and general-purpose-AI enforcement remain live from 2 August 2026 regardless, so the near-term deadline hasn't actually moved for most organizations. Holistic AI's own coverage of this gets the nuance right, including noting that ISO/IEC 42001 and the NIST AI RMF are 'a genuinely useful foundation' but don't substitute for the Act's own conformity assessment, CE marking or EU database registration — a more careful framing than most of the 'AI Act delayed' coverage that circulated this year.

## Holistic AI price and terms

### Terms

- Free to start: Limited free use No free or trial tier of the governance platform is published
- Sign up without sales: No

The terms in full

the free path is the separate Apache 2.0 holisticai Python library, which measures bias, explainability, robustness, privacy and efficacy locally and does nothing the platform does for discovery, enforcement or audit evidence.

### For AI agents

- API: Partly [holisticai.com, 28 October 2026](https://www.holisticai.com/enforce)

[![Holistic AI](https://agentsindex.ai/img/1024/screenshots/holistic-ai/pricing.webp)](https://agentsindex.ai/img/1920/screenshots/holistic-ai/pricing.webp)

Pricing page, captured 21 October 2026

[holisticai.com](https://www.holisticai.com/), checked 28 October 2026

## What we checked on Holistic AI

Yes Partly No

What it costs to start Yes

### What the price counts

Pricing is quote only: no rate card, plan table or figure appears anywhere on the site, and /pricing returns 404, so every buyer is routed to a sales conversation.

> Join the organizations that turned governance from a blocker into an enabler. Full visibility, continuous risk testing, and compliance proof — on…

Closing call-to-action block, whose only action is Get a Demo [holisticai.com](https://www.holisticai.com/) Checked 28 October 2026

What it costs to start Partly

### Free to use

The commercial governance platform cannot be used without a sales conversation, but the vendor publishes a free Apache 2.0 Python library for bias, robustness and privacy measurement.

> An open-source library for assessing and improving the safety aspects of AI systems.

README, opening description [github.com/holistic-ai/holisticai](https://github.com/holistic-ai/holisticai) Checked 28 October 2026

What it costs to start No

### Self-serve signup

Every call to action on the platform, product and integrations pages ends at a demo request form; the site publishes no sign-up, trial or self-checkout path.

> Need a custom integration? Our platform connects with any tool in your stack. Let our team help you get set up.

Closing block of the integrations list, where the only action offered is Contact Us [holisticai.com/integrations](https://www.holisticai.com/integrations) Checked 28 October 2026

What it works with Yes

### Works with

Named read-only connectors cover GitHub, GitLab, Bitbucket, SharePoint, Confluence, Azure, GCP, AWS, Snowflake, Databricks, ServiceNow, Langsmith, Langfuse, CrewAI, Copilot Studio and Zscaler, with no plan gating stated.

> Holistic AI seamlessly integrates with your existing tools and platforms to provide comprehensive AI discovery and governance across your entire…

Integrations hero, above the connector grid [holisticai.com/integrations](https://www.holisticai.com/integrations) Checked 28 October 2026

What it works with Partly

### Public API

The vendor states that enforcement is deployed through an API or SDK and lists REST APIs, webhooks and SDKs as connection options, but publishes no public API reference for the platform; the readthedocs site documents the separate open-source Python library.

> Activate enforcement across models, agents, APIs, and workflows in real time via API or SDK.

How Enforce Works, step 02 Deploy guardrails [holisticai.com/enforce](https://www.holisticai.com/enforce) Checked 28 October 2026

What it does on its own Yes

### Acts on its own

Operative Guardian Agents act without a person approving each step: they block requests, revoke privileges, quarantine agents and trigger kill switches once a defined risk threshold is crossed.

> When risk crosses a pre-defined threshold, Operative Agents step in automatically. They block, redirect, quarantine, or remediate unsafe AI behavior…

Two Modes of Protection, Operative Agents panel [holisticai.com/guardian-agents](https://www.holisticai.com/guardian-agents) Checked 28 October 2026

What it does on its own Yes

### Limits on what it may do

Limits are set in advance as policies and thresholds: a visual builder, regulatory templates and a custom rule engine define what is allowed before any Guardian Agent intervenes.

> Create and customize AI policies using a visual builder or pre-built regulatory templates.

The Solution, step 01 Define, Policy Configuration [holisticai.com/enforce](https://www.holisticai.com/enforce) Checked 28 October 2026

What it does on its own Partly

### No code needed

Policies are built and tested without code through a drag-and-drop builder and templates, but connecting sources and deploying runtime guardrails is described as an API or SDK task.

> Create policies with drag-and-drop interface

Capabilities, Policy Management tab [holisticai.com/enforce](https://www.holisticai.com/enforce) Checked 28 October 2026

What happens to your data Yes

### No training on your data

The vendor states flatly that customer data is never used to train models and that it trains no models of its own; the platform only governs systems the customer builds or buys.

> No. We do not use customer data to train models.We do not build or train models of our own.

FAQ, Data handling, 'Do you train models on customer data?' [holisticai.com/trust-center](https://www.holisticai.com/trust-center) Checked 28 October 2026

What happens to your data Partly

### Retention you control

Retention is bounded by the purpose or the timeline written into the customer MSA and deletion can be requested at any time, but no page offers a self-service retention setting in the product.

> Data is retained only as long as necessary for the purpose it was collected for, or according to the timeline set out in the MSA.

Security programme, Data and privacy, Retention and deletion control [holisticai.com/trust-center](https://www.holisticai.com/trust-center) Checked 28 October 2026

What happens to your data Yes

### Data stays in your region

Hosting location follows the customer: UK by default on AWS Europe (London), United States customers on AWS US East (Ohio) or Google Cloud where required, with transfers covered by the DPA.

> Customer data is hosted in the UK by default. US customers are hosted in the United States.

Security programme, Data and privacy, Data residency control [holisticai.com/trust-center](https://www.holisticai.com/trust-center) Checked 28 October 2026

What happens to your data Yes

### Open source

Apache-2.0 licence in its public repository

> Apache License Version 2.0

LICENSE in github.com/holistic-ai/holisticai [github.com/holistic-ai/holisticai/blob/…](https://github.com/holistic-ai/holisticai/blob/main/LICENSE) Checked by us, 15 November 2026

What happens to your data Yes

### Take your data out

Audit evidence and compliance reports export with full traceability, and governance data can be synced out to Snowflake; deletion on exit is confirmed in writing, though no bulk account export is described.

> Export audit-ready evidence with complete traceability.

How Enforce Works, step 04 Generate reports [holisticai.com/enforce](https://www.holisticai.com/enforce) Checked 28 October 2026

Whether a company can buy it Yes

### Security audits

ISO/IEC 27001:2022 certification with 92 of 93 Annex A controls in scope and a SOC 2 Type 2 report covering four criteria; ISO/IEC 42001 is stated as not yet held.

> Information security management system supporting the development and deployment of the Holistic AI platform.

Compliance section, ISO/IEC 27001:2022 card marked Certified [holisticai.com/trust-center](https://www.holisticai.com/trust-center) Checked 28 October 2026

Whether a company can buy it Yes

### Single sign-on

Single sign-on and user directory management are part of platform sign-in, run through WorkOS, which also appears as a named subprocessor; no plan gating is stated.

> Platform sign-in, single sign-on and user directory management run through WorkOS.

Security programme, Access control, Secure authentication control [holisticai.com/trust-center](https://www.holisticai.com/trust-center) Checked 28 October 2026

Whether a company can buy it Yes

### Audit log

The platform keeps full audit trails of AI system changes, model versions and agent activity, with decision logging generated automatically for compliance evidence.

> Full audit trails for all AI system changes, model versioning, and real-time agent observability.

Platform section, Observability & Audit card [holisticai.com/identify](https://www.holisticai.com/identify) Checked 28 October 2026

## Questions about Holistic AI

- How much does Holistic AI cost?: Holistic AI gives its price on request. From holisticai.com, checked 28 October 2026.
- Does Holistic AI have an API or an MCP server?: Holistic AI has a limited API. From holisticai.com/enforce, checked 28 October 2026.
- Is Holistic AI open source, and can you host it yourself?: Holistic AI is open source. We checked github.com/holistic-ai/holisticai/blob/main/LICENSE ourselves on 15 November 2026.
- What have we checked on Holistic AI?: 16 facts, under these headings: What it costs to start, what it works with, what it does on its own, what happens to your data, whether a company can buy it.
- How do we check Holistic AI?: We hold 16 facts on Holistic AI, quoted from 8 pages. Facts last checked 15 November 2026. Price last checked 28 October 2026. Last change found 15 November 2026.

[How we check every listing](https://agentsindex.ai/methodology)

## Sources and changes

Last full check 28 October 2026

Sources 7

- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/trust-center) /trust-center
- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/enforce) /enforce
- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/) /
- Read 28 October 2026 · [github.com](https://github.com/holistic-ai/holisticai) /holistic-ai/holisticai
- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/integrations) /integrations
- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/guardian-agents) /guardian-agents
- Read 28 October 2026 · [holisticai.com](https://www.holisticai.com/identify) /identify

What changed 1

- 15 November 2026 What it does

## Holistic AI price history

We have read the price once so far, on 28 October 2026: price on request.

| Price | Plan | First seen | Last seen | Readings | Source |
| --- | --- | --- | --- | --- | --- |
| Price on request |   | 28 October 2026 | 28 October 2026 | 1 | [Pricing page, read 28 October 2026](https://www.holisticai.com/) |

From our research on 28 October 2026. A date is when we read the page, not when the price changed.

## Similar to Holistic AI

- [Base2026](https://agentsindex.ai/base2026) Free SEO research MCP server, public API and browser SEO checks with original sources attached. Free plan
- [CurlShot](https://agentsindex.ai/curlshot) Screenshot and PDF rendering API with an MCP server, so an agent can see any web page. Free plan, then $19 a month
- [Bot Street](https://agentsindex.ai/botstreet) Marketplace where AI agents post offers, take paid tasks and run shops, reached over MCP, a CLI or REST. Free plan

## Is Holistic AI yours?

Free, and the page then carries the verified owner mark. You confirm it with an email at holisticai.com. [Claim this listing](https://agentsindex.ai/claim/holistic-ai?from=owner-line)

[Report an error](mailto:hello@agentsindex.ai?subject=Report%20an%20error%3A%20Holistic%20AI%20(holistic-ai)&body=What%20is%20wrong%20on%20the%20page%2C%20and%20where%20on%20the%20product) if a fact is wrong.
