Ask a question. See exactly how it was answered.
Argos Data Agent turns a plain-language question into a governed query against the reports and DataBlocks you already trust, and shows its work every time. The question that used to wait days on a report request gets answered the moment it is asked, inside the permissions IT already set.
Built on the governed reporting layer institutions have trusted for two decades
What is Argos Data Agent?
AI that answers from data your institution already governs
Ask a question the way you would ask a colleague, and get an answer you can take into a dean's office. Anyone with Argos can do it, in seconds, without SQL and without filing a request. Your team decides which DataBlocks it is allowed to answer from.
The answers come from the reporting your team already built and already trusts, so the number agrees with everything else the institution publishes. Each one shows how it got there, and nobody sees anything they were not already entitled to see.
If you run Argos, there is nothing to build first. It answers from the DataBlocks your team has already written, with the joins, filters, and definitions already settled, so the first question works against the reports you have today.
The situation
AI is already touching institutional data, and nobody logged it
Policies are lagging adoption. EDUCAUSE research found practitioners nearly evenly split on whether they are even aware of guidance covering their own work-related AI use, and reported that institutions have more AI tools running across their campuses than IT departments know about.
Knowledge management for safer AI ranks fifth on the EDUCAUSE Top 10, framed as mitigating AI risk by integrating knowledge management into data governance, privacy, and ethics programs. Governance comes first, then the tool.
01
Your staff are already asking AI about institutional data
Someone exported a spreadsheet into a general assistant and asked it a question. It was fast, it needed no approval, and nobody logged it.
02
The risk is a plausible number, not an obvious error
A model working from stripped context, definitions it never saw, and permissions nobody checked will still answer. Then that number goes into a deck.
03
Verification is what makes an answer usable
In a place where figures get quoted to boards, agencies, and legislators, an answer nobody can reproduce is not an answer.
04
Nobody has capacity for a data project first
Most AI analytics proposals start with a warehouse, a semantic layer, and a full semester before anyone asks a question. Data Agent skips that stage. It starts from the reporting you already have.
Governed DataBlocks
It queries what your team defined, not your raw database
Data Agent does not translate a question into SQL against your ERP schema. It works from DataBlocks where the joins, filters, and institutional definitions are already settled.
A model pointed at a raw database schema has to infer what your tables mean, and it will infer wrong without telling you. A model querying objects your team already defined is working from settled meaning, and every answer shows the DataBlock and the SQL it used, so a wrong one can be caught by the person who owns the definition.
- Answers grounded in Argos DataBlocks your institution already stands behind
- Your definition of full-time, census, or a cohort, rather than the model's guess
- Works with the DataBlocks you have already built, with nothing to move or rebuild first
SELECT channel, cohort, retainedFROM DataBlock: Admissions funnelWHERE term = :termExplainability
Every answer shows its work
Data Agent exposes the SQL it ran and the DataBlock it ran against, beside every response. An analyst can read it, reproduce it, and confirm the cohort definition matches the certified one before anything acts on it.
The same answer serves the committee and the person who has to defend it.
- The SQL it ran and the DataBlock it ran against, shown with every response
- An analyst can reproduce any answer independently
- Reduced risk of misinterpretation, and no black box to defend to a committee
- Trusted by technical and non-technical users for the same reason
Beyond retrieving a number
The second and third question, without starting over
Your reports already retrieve numbers. The harder problem is that a result with two thousand rows can contain the answer without making it obvious, and reading it is a skill not everyone has.
Data Agent summarizes large result sets, compares populations and time periods, and answers the follow-up question in the same conversation, without rebuilding anything. At the Advanced tier it reasons across more than one DataBlock and can dig into what is contributing to a shift when you ask it to.
- Summarize a result of hundreds of thousands of records and say what changed
- Compare programs, departments, terms, and cohorts against each other
- Ask what is driving an enrollment shift or a retention change, and get an answer with the query behind it, at the Advanced tier
- Build a chart from any answer
- Reason across more than one DataBlock in a single conversation, at the Advanced tier
- Keep going without filing a new reporting request
Governance
The controls your policy committee is going to ask about
- Zero Data Retention, with encryption in transit
- MAPS permissions honored, so there is no second access model to administer
- Data Masking Manager: block a DataBlock, mask a column, redact a value
- Cloud, hybrid, and on-premises deployment all supported
- One HECVAT and one Trust Center covering the platform
What is included
Two tiers, built for your needs
Both tiers run on the same governed DataBlocks, the same MAPS permissions, and the same Zero Data Retention. The difference is reach and depth.
| Core | Advanced | |
|---|---|---|
| Who it is for | Everyone who needs faster answers from approved reports | Analysts, IR, ITS, report owners, and power users |
| Scope of a question | One governed DataBlock | Multi-DataBlock reasoning and cross-domain analysis in one conversation |
| Where it opens from | Explore Report Data, on any existing report | Any folder or dashboard across Argos X |
| What it answers | Questions about the report or DataBlock you are looking at: summaries, totals, and comparisons within that dataset | Deeper analysis across DataBlocks: trends, recommendations, and rich data exploration |
| Chart creation | Yes | Yes |
| Relationship analysis | No | Yes |
| AI-ready Starter Pack DataBlocks | Yes | Yes |
| SQL shown with every answer | Yes | Yes |
| MAPS permissions honored | Yes | Yes |
| Data Masking Manager | Yes | Yes |
| Zero Data Retention | Yes | Yes |
| Licensing | Campus-wide subscription, unlimited users | Seat licenses your admins assign |
Argos Data Agent is licensed separately from Argos X, which includes Dashboard Designer, Explore Data, and Pivot Mode at no additional cost.
An AI answer your governance committee can verify
Every response shows the SQL it ran and the DataBlock it ran against, and reads only what Argos permissions already allow. See it answer a question in a demo.
Case study
1 day
to set up on existing DataBlocks
240
trained users with Data Agent access
~15 seconds
to break down grades across 2,000+ graduates
Texas A&M put governed student data within reach of 240 users
“When it gives me the SQL statement, I go look at the where clause and see how it got there. It’s nice that it explains how it got to that point, not just here’s the question, here’s the answer.”
Josh Radnitzer · IT Manager IV, Enterprise Information Systems, Texas A&M
Related
Where the governed layer comes from
Argos
The reporting layer Data Agent answers from, and how DataBlocks get built.
AI for Higher Education
The governance argument, and where AI on institutional data goes wrong.
Platform and Security
MAPS, permissions, deployment, and what your reviewers will ask.
Reporting for IT
Access control, the report queue, and who gets to answer questions.
How it fits
Where this sits relative to what you already run
With Argos
Data Agent is not a separate system. It answers from the DataBlocks, permissions, and definitions your Argos environment already maintains, so there is no data project to run before the first question.
With a general-purpose AI assistant
A general assistant drafts and summarizes well and knows nothing about your data model. Point one at a student system and it has to guess what your tables mean. The two are complementary and should not be asked to do each other's jobs.
With Ellucian Insights
Insights aggregates data across Ellucian products on a managed data layer. Argos works directly against your own environment, covering dashboards through Dashboard Designer, conversational analysis through Argos Data Agent, compliance output, and scheduled delivery, on definitions your team controls. Most institutions run both.
Trust and security
You take security seriously. So do we.
The Trust Center is published and the HECVAT is available on request, so a reviewer can start before anyone schedules a call.
Zero Data Retention
Institutional data submitted through the AI workflow is not retained by Evisions or by the model provider, for training or for anything else.
Permissions are enforced, never overridden
MAPS permissions are enforced, so a question cannot reach a DataBlock the person asking is not permissioned to run. There is no parallel access model to administer.
Data Masking Manager
Block an entire DataBlock from AI workflows, mask specific columns, or redact specific values by exact match or pattern. The settings apply to everyone who uses the agent.
Verifiable answers
The SQL it ran and the DataBlock it ran against, shown with every response, so an answer can be audited instead of taken on trust.
Deployment
Cloud, hybrid, and on-premises all supported, with encryption in transit.
Documentation
Trust Center at trust.evisions.com, and a HECVAT on request.
Frequently asked questions
Questions we get asked first
Is our data used to train an external model?
No. Zero Data Retention means institutional data submitted through the AI workflow is not retained by Evisions or by the model provider for training or reuse.
How do we verify an answer before acting on it?
Every response exposes the SQL, the applied logic, and the data sources. An analyst can read it, reproduce it, and confirm it.
Can we restrict which data the AI can reach?
Yes, at three levels. Entire DataBlocks can be blocked from AI workflows, columns masked, and values redacted by exact match or pattern, with your existing permissions applying on top. The settings apply to everyone who uses the agent.
Does this create a second access model?
No. It enforces the MAPS permissions you have already configured, so there is nothing to keep in sync.
Do we need a data warehouse or a semantic layer first?
No. It works on the Argos DataBlocks you already have. If you do not run Argos yet, start there rather than here.
How accurate is it?
We publish the reasoning instead of a score. Every answer arrives with the SQL beside it, so your own team can confirm the figure instead of trusting a percentage we chose. Bring the questions your office asks and we will run them against your data.
What is the practical difference between Core and Advanced?
Reach and depth. Core answers from one governed DataBlock and opens from Explore Report Data on any existing report. Advanced reasons across multiple DataBlocks in a single conversation, opens from any folder or dashboard in Argos X, and adds relationship analysis. Both tiers build charts. Security controls, masking, and permissions are identical.
How is each tier licensed?
Core is licensed per institution with unlimited users. Advanced is licensed per seat. Tell us who would be asking questions and we will tell you which tier fits.
Will this replace our analysts?
No, and we would be cautious of anyone claiming it will. It reduces repetitive retrieval so analysts keep ownership of governance, validation, and the questions that need judgment.
Who is this a good fit for?
Institutions already running Argos with DataBlocks containing real institutional logic, where the reporting team gets more requests than it can work through and leadership expects a controlled AI use case. If that does not describe you yet, the DataBlocks are the part to get right first.
Can we turn it off, or limit who can use it?
It depends on the tier. Data Agent Core is licensed per institution, so once it is turned on, anyone with access to your Argos environment can use it. Data Agent Advanced is licensed per seat, so you decide who gets access by assigning seats. For both, what you control is what the agent can reach: you choose which DataBlocks are available to it. If you do not want it answering questions about finance data, you leave the finance DataBlocks out and it cannot answer them for anyone.
Who will we be working with?
People who have done your job. Our consultants and support staff came from campus IT and institutional reporting before they worked here, because higher education is the only sector we build for. Support is in-house, with no outsourced agents, and it is the thing customers raise unprompted more than anything else about us.
How long does implementation take?
If Argos is in place, this is a license and a configuration instead of a project.
What does it cost, and can we talk to an institution using it?
Core is per institution with unlimited users, Advanced is per seat, and price is driven by institution size and which tier fits. We will give you a real figure on the first call. Ask about references and we will tell you where the deployment base stands today.
Next step
AI your institution can say yes to
Plain-language answers, visible queries, and the permissions IT already set. Get a demo of Argos Data Agent.
Ellucian Partner Award
2025: Customer First
Ethos Connected
since 2020
Ellucian partner
22 years