Company Values

Ethical AI and the Ethical Use of Technology

At Iternal Technologies, responsible AI is a collective responsibility. We keep humans in the loop, comply with the EU AI Act and enterprise standards, and steward your data with AI that runs local and secure — upholding human dignity and fundamental rights for all people.

What Is Ethical AI?

Ethical AI is the practice of building and operating AI systems that respect human rights, the law, and the people affected by their outputs. It rests on six principles: fairness, transparency, accountability, safety and security, privacy, and meaningful human oversight. The test is what a system does in production, not what a policy promises.

Most of the difficulty is architectural rather than philosophical. An AI assistant that ships a confidential document to an external service to answer a question about it has already made a privacy decision on the organization's behalf, whatever the policy says. Iternal's answer is to remove that decision from the runtime: AirgapAI runs the model on the same machine as the data, and Blockify keeps every answer tied to the source document it came from.

The governing structure that turns these principles into approvals, risk tiers and review gates lives on the AI governance framework page. This page states what Iternal holds itself to and how each principle is enforced in the products it ships.

"Technology has the incredible power to change the world. It's how we use it that determines if the impact is good or bad. At Iternal, we believe in building technology that can be used to positively change the world and better society for generations to come."

John Byron Hanby, IV — Founder & Chief Executive Officer, Iternal Technologies

Responsible AI in Practice

How we build and deploy AI that respects human judgment, the law, and your data

Human-in-the-Loop

We build AI that augments human judgment, not replaces it. People stay in control of high-stakes decisions, and our systems are designed for transparency so teams understand how outputs are produced.

EU AI Act Aligned

Our platform is built to comply with the EU AI Act alongside HIPAA, CMMC 2.0, ITAR, and GDPR — governance by design so organizations can deploy AI responsibly in regulated environments.

Data Stewardship

Data protection is foundational: AirgapAI applications run 100% on local hardware, so sensitive data never leaves your perimeter. We minimize what we collect and keep you in control of your information.

Our Collective Responsibility

Over the last 150 years, the number of technological milestones that shape our world today are simply incredible. There has been so much positivity that has come from innovation in technology, but we also recognize the massive responsibility we as the human race bear in responsibly creating and developing technologies.

It's our collective responsibility to ethically use technology.

We believe the technology and innovations we develop should always be used in an ethical manner that upholds human decency and the fundamental human rights granted to every living person. We know that ethical and responsible use of our technology will lead to a better world for everyone, and by extension, more success for our customers.

We take great care, to the best of our abilities, to ensure our technology and the implications of our technology are clearly thought through and discussed both internally within Iternal and externally with experts.

Responsible AI Principles and How Iternal Enforces Each One

Six principles, and the mechanism in AirgapAI and Blockify that makes each one hold at runtime

01

Fairness

AI systems should treat people equitably, and bias in training data, retrieved content or outputs has to be found deliberately rather than assumed away.

How it is enforced: Blockify turns source material into IdeaBlocks a subject-matter expert can read, approve or delete, so the content a model answers from is a reviewed dataset instead of an opaque index. Organizations choose the models AirgapAI runs and can test them on their own material before a high-risk use case is approved.

02

Transparency

People should know when AI was involved, where an answer came from, and what the system cannot do.

How it is enforced: Every Blockify IdeaBlock links back to its source document, so answers arrive with provenance a reviewer can follow to the page it came from. Disclosure of AI use is written into policy through the AI acceptable use policy rather than left to individual judgment.

03

Accountability

Every production AI system has a named owner who is answerable for what it produces and what it costs.

How it is enforced: Deployments are tiered by risk before approval, and each tier names its approver and the validation evidence required. Iternal engagements assign that owner during the pilot rather than at rollout, when the system is already in front of customers.

04

Safety and Security

AI systems should be engineered to prevent harm, resist attack and contain failure at the application, model, data and infrastructure layers.

How it is enforced: AirgapAI runs the model, the data and the retrieval step entirely on the customer’s own hardware, with no external calls, which is why it operates offline and inside SCIFs. It is deployed against HIPAA, CMMC/NIST, FedRAMP and GDPR data-residency requirements; see the AI compliance frameworks overview for the mapping.

05

Privacy

Personal and confidential data should be minimized, kept inside the boundary that already governs it, and never repurposed without consent.

How it is enforced: With AirgapAI the data never leaves the machine it already sits on, so there is no external copy to subpoena, breach or retain. Blockify and AirgapAI do not train models on customer data, and the EU AI Act obligations that fall on a deployer can be met inside the customer’s own perimeter.

06

Human Oversight

Human control must be meaningful and scaled to consequence, from a light review on low-risk drafting to mandatory sign-off on decisions about people.

How it is enforced: Oversight is designed in tiers rather than declared once: AI carries the volume, a qualified person owns the commitment, and nothing consequential is finalized or sent externally without that review. The design is documented in the 70-30 human-in-the-loop model.

Our Approach to Ethical Decisions

We strive to have meaningful and open discussions with our employees, customers, and partners. Hearing a variety of perspectives and having productive discussions allows us to do our best to answer and create a path forward for really complex decisions.

We also rely on the advice and guidance from experts in both corporate and academic circles to help our teams create a framework for responsibility.

Employee Engagement

Open discussions with team members at every level

Customer & Partner Input

Gathering diverse perspectives from our ecosystem

Expert Consultation

Corporate and academic guidance on complex issues

Framework Development

Creating structured approaches to responsibility

Our Commitments

How we ensure ethical practices guide every decision we make

Thoughtful Internal Discussions

We prioritize deep, thoughtful internal discussions about the implications of our technology before it reaches the market.

External Consultation

We actively consult with corporate and academic experts to ensure our innovations meet the highest ethical standards.

Open Dialogue

We maintain open dialogue with employees, customers, and partners to understand diverse perspectives on technology impact.

Multiple Perspectives

We seek multiple perspectives on complex ethical decisions, recognizing that responsible innovation requires diverse viewpoints.

Why Ethics Matter

Consumer loyalty is directly tied to ethical business practices

86%
Show greater loyalty to ethical businesses
75%
Would avoid purchasing from unethical companies
69%
Spend more with responsible organizations

Source: Salesforce Research, October 2018

Addressing Consumer Concerns

We take seriously the concerns that consumers have about technology

93%
Worry about misinformation
91%
Concerned about surveillance
88%
Fear job displacement
85%
Worry about democratic threats
83%
Concerned about environmental damage

The Path Forward

Consumers believe technology can be a force for good

67%
Believe technology itself is neutral — it's the implementation that determines impact
81%
Believe emerging technology can improve society when used responsibly

Responsible AI FAQ

Common questions about how Iternal builds and deploys AI responsibly

Iternal builds AI that augments human judgment rather than replacing it. People stay in control of high-stakes decisions, and our systems are designed for transparency so teams understand how outputs are produced.

Yes. Iternal’s platform is built to comply with the EU AI Act alongside HIPAA, CMMC 2.0, ITAR, and GDPR, with governance by design so organizations can deploy AI responsibly in regulated environments.

Data stewardship is foundational. AirgapAI applications run 100% on local hardware, so sensitive data never leaves your perimeter. Iternal minimizes data collection and gives you full control over your information.

Responsible AI is the operating discipline that puts ethical intent into practice: deciding which AI systems are allowed to run, tiering them by the consequence of getting them wrong, naming an owner for each one, keeping data inside the boundary that already governs it, and requiring human review in proportion to the risk. Ethics supplies the principles; responsible AI is the set of controls, reviews and architecture choices that make those principles observable in a running system.

Ethical AI is the standard — fairness, transparency, accountability, safety and security, privacy and human oversight — and it answers whether a system should exist in the form proposed. Responsible AI is the implementation of that standard: the approvals, risk tiers, review gates, data boundaries and audit trails that make it hold in production. The terms are used interchangeably in most coverage, and the practical distinction is that one is judged by intent and the other by evidence.

Four things, taught with examples from the work people actually do: when to disclose that AI was used, how to verify a cited source before acting on an answer, which data may be put into which tool, and the decisions that always require a person to sign off. Iternal AI Academy delivers this as course-based training ending in a certificate of completion, and the rules it teaches are written down in an AI acceptable use policy so they outlast the session.

AI Ethics Training for Employees

Principles hold when the people using AI every day know what they mean in their own work

Course-based training

Iternal AI Academy teaches the working habits behind these principles: disclosing AI use, checking a cited source before forwarding an answer, and recognizing the tasks where a person has to stay in the loop. Courses end in a certificate of completion.

Iternal AI Academy

Rolled out to a whole workforce

Ethics training only changes behavior when every team gets the same version of it, in the same quarter, with the same examples from their own work. That program view — cohorts, role-based tracks, measurement — sits on the employee training page.

AI training for employees

Written down where people look

Training fades; a one-page policy that names what is allowed, what needs approval and what is never permitted survives the quarter. The acceptable use policy is the artifact the training points back to.

AI acceptable use policy

Have Ethical Concerns?

We welcome open dialogue about the ethical implications of our technology. Reach out to discuss how we can work together responsibly.

Contact Us
(function () { document.querySelectorAll('.ethics-faq-q').forEach(function (btn) { btn.addEventListener('click', function () { var item = btn.closest('.ethics-faq-item'); if (!item) return; var isOpen = item.classList.toggle('open'); btn.setAttribute('aria-expanded', isOpen ? 'true' : 'false'); }); }); })();