How we use AI, and where we don't.

We help companies use AI responsibly, so it's only fair to show the rules we follow ourselves. This page covers the website, the images and video on it, what we publish, and how we handle AI in client work.

On this website

• This website makes no live calls to any AI model. Its pages are static files, built in advance.
• The AI Readiness Assessment and the Automation Payback Calculator run in your browser, using fixed questions, rules and formulas. No AI model is involved, and nothing you type reaches us unless you choose to include a result in a message.
• Nothing on this site uses AI to profile visitors or to change what you see.

In the images and video on this site

They show the kind of work we do: inspection benches, production lines, documents and instruments. They are not photographs of EliteWorks premises, clients, founders, specialists or delivered projects, and we never present them that way.

Every image and video is checked before it goes live. It must contain no readable text, no logos or brand names, no recognisable real person, and no weapons or defence systems. Where people appear, they are hands or figures nobody could identify, and they are never captioned as anyone who works with EliteWorks.

When photographs of the founders are added, they will be real photographs supplied by the founders themselves.

In what we publish

Our guides give our own findings and judgement. AI tools may help organise notes or edit drafts. The named author checks every fact, figure and source, and no guide goes live until its author has reviewed it. We don't publish statistics we can't trace to a named source, and we don't put words in the mouths of people who didn't say them.

In client work

We never put a client's confidential data into a third-party AI tool without the client's written agreement, saying which tool, for what purpose and where the data is processed. Where a client's policy says the work must stay inside its own systems, it does.

Any AI system we build for a client follows the same standards we recommend to others: pass criteria written down before the build, tests built from real cases, limits set action by action on what the system may do alone, a person's approval for any action with real consequences, release in stages, and decisions logged. See how we work for the detail.

Questions

If you want to know more about how we use AI, including for due diligence, write to support@eliteworks.ai. A founder replies personally within one business day, Monday to Friday, India time.

Version history. responsible-ai-v1-2026-09: first version.