Responsible AI for Organizations.Powerful AI for Practitioners.
Justice AI GPT helps organizations govern AI responsibly while empowering educators, researchers, coaches, consultants, and social impact professionals with an AI assistant built for ethical, accountable, and high-quality work.
Built for trust. Designed for governance. Created for people who need AI they can stand behind.
Justice AI GPT
The world's first Decolonial AI Framework.
Bias is not only a problem of language. It can be embedded in assumptions, institutional norms, decision rules, and definitions of whose knowledge is treated as authoritative.
The response reproduces several patterns that warrant examination.
⚖️ Run Bias AuditReviews risk before it becomes practice.
Every AI decision becomes someone's reality.
- HR and hiring decisions
- Healthcare communication
- Education and curriculum
- Policies and procedures
- Public communications
- AI content and workflows
- Internal governance
- Responsible AI adoption
In 2025, Justice AI GPT identified an estimated $1.2 billion in potential legal, compliance, and operational risk exposure.
Protect your work. Start your membership
Two ways to work with us
One framework. Two separate doors.
Practitioners use Justice AI every day. Institutions commission independent audits. These are different services from different organisations, and we keep them apart on purpose.
For practitioners
Justice AI is the working tool: the decolonial dictionary, the document library, structured analysis and built-in bias review. Educators, researchers, coaches, consultants and organisers use it day to day.
For institutions
The Coalition audits how institutions use AI, make decisions and share power, and helps build the governance that fixes what it finds. Fees are published, and any adverse finding goes to an outside reviewer.
Liberation is the standard. Human rights are the floor. Institutional accountability is the obligation.
From the DEIBA, Intersectional Human Rights & Liberation Audit and Governance Standard, Version 1.0.
How an audit works
Four questions. Every audit.
Every engagement answers the same four questions, in this order, and publishes what it found.
What harm is happening?
We look at whether a rule, a form, a fee, a contract, a model or a decision is causing discrimination, exclusion or unfair treatment, and how that harm became routine.
Who carries more of it?
People facing more than one form of harm at once often carry the most and get counted the least. We name the groups we looked at, what we compared them with, and where the evidence ran out.
Who keeps it in place?
We name who approves the structure, who profits from it, who maintains it and who pays the cost. A diverse committee or an inclusion policy does not answer this question.
What has to change, and did it?
Each finding names who is responsible, what must be corrected and the date we check it. A fix that leaves the power where it was is named as exactly that.
Independence and disclosure
Same founder. Separate organisations. Said out loud.
You cannot credibly audit AI governance while selling the AI platform under review. So the Decolonial Algorithmic Coalition does the audits and governance building, and Justice AI GPT, LLC sells the platform. Both were founded by Christian “ZacaTechO” Ortiz, and they are legally separate.
Any use of a Justice AI product inside an engagement is disclosed and priced separately before it begins, and any finding that touches the product goes to the outside reviewer. No audit requires a Justice AI licence, and holding one never changes the fee.
Pricing
Published fees. A floor for the organisations that need it most.
Fees are set by tier, published in US dollars, and never change with what an audit finds. Most organisations start with a three-month pilot, then move to an annual programme.
| Tier | Who it covers | One audit | Annual programme |
|---|---|---|---|
| Community | Grassroots groups, mutual aid networks, Indigenous-led and community organisations, 1–3 staff | $12,000–$20,000 | $52,000–$85,000 |
| Mission | Nonprofits, schools and small public agencies, 4–250 staff | $30,000–$55,000 | $135,000–$235,000 |
| Institutional | Cities and counties, hospital systems, universities and mid-size companies, 251–5,000 staff | $65,000–$120,000 | $290,000–$520,000 |
| Enterprise | Over 5,000 staff, or any organisation whose automated decisions decide access to housing, credit, care, work or education at scale | $140,000–$260,000 | $620,000–$1,100,000 |
Community organisations with a budget under US$500,000 pay from $6,000. Institutional and Enterprise fees are the same in every country. Publication is never a payment milestone.
What you receive
Documents you can act on and defend.
A written scope
The systems, policies and time period under review, agreed and signed before any fieldwork.
Findings, sorted honestly
Proven harm, unresolved risk, not enough evidence, or out of scope. No finding without a stated basis.
A named owner
The office or role responsible for each correction, not just a department.
A second set of eyes
An outside reviewer on every adverse finding, with a window to respond and a published decision.
A follow-up check
We check whether each correction happened and report it in writing, whether it did or not.
Published corrections
Factual corrections are published, dated and linked from the original finding.
Who the Coalition works with
We take on
Cities, counties, hospital systems, universities, foundations, nonprofits, school districts, and platforms whose automated decisions decide who gets housing, credit, care, work or education.
We decline
Work designed to produce a predetermined finding or hide proven harm, and work that would lend legitimacy to coercive surveillance, carceral harm, militarized oppression or exploitative systems.
Refusals follow published criteria about a practice and its effects, never assumptions about an organisation's identity or reputation.
Start with the standard. Then ask us anything.
Read the standard first. If it holds up for your institution, request a scoping conversation. We will tell you plainly if an audit is the wrong tool for what you are facing.

