Back to blog
AI & BusinessSeptember 22, 20264 min readBy Kwata Team

Five questions to ask before you trust any tool with your work

Five questions to ask before you trust any tool with your work

When you let a new tool into your work, you are trusting it before it has earned that trust.

You sign up and upload a file, and it quietly becomes part of how you run your practice, your business or your organisation. The trouble is that most of the real answers are buried in pages nobody reads, and you usually only find out what a tool does with your work when something goes wrong.

You do not need to be technical to protect yourself. You need to ask five plain questions, and to notice how a tool answers them.

1. Does it tell you what it cannot do?

Good tools are honest about their limits. They tell you when they are unsure, when they could not check something, and when you should look again yourself.

Be wary of a tool that sounds equally confident about everything. Confidence is easy to build into software. Honesty takes deliberate work, and it is the thing that protects you when the answer matters.

Ask: when this tool does not know, how does it tell me?

2. Where does your data go, and does it train anyone's AI?

This is the question most people forget to ask, and the one that matters most. Your files, your clients' details and your numbers are valuable, and some services use what you give them to improve their own products.

Look for a plain answer, in writing, to two things. Is your data ever used to train AI? Can you export it or delete it whenever you choose?

Ask: if I leave tomorrow, what happens to everything I put in?

3. Is it watching you when you are not using it?

Many websites follow visitors around the internet with advertising trackers, so an ad for the thing you glanced at follows you for weeks. That is annoying on a shopping site. On a tool that handles confidential work, it tells you something about how the company treats attention and data.

Ask: does this company track me across the web, or only measure how its own site is used?

4. Are its badges real?

"Certified", "compliant" and "bank-level security" are easy words to put on a website. Some are backed by an independent audit. Many are not.

There is an honest version, and it is easy to spot: a company that tells you which standards it is built to, and which formal audits are still on its roadmap. That company is telling you the truth about where it stands. That matters more than any badge.

Ask: which of these claims has an independent auditor actually signed?

5. Does it leave the decision with you?

The best tools make you sharper. They set out the facts, the risks and the options, then step back. A tool that tries to make the call for you, or presents a guess as a certainty, is asking you to hand over judgement that should stay yours.

Ask: does this tool show me its reasons, or just give me an answer?

What good answers look like

At Kwata Team, these are the questions we ask of our own work, whether that is Juris for law firms, Kwata Books for small businesses, or a website or custom web app we build for an organisation.

Your data is never used to train AI and never sold, and you can export or delete it at any time. The Juris site carries no advertising or retargeting cookies. Our security controls map to the SOC 2 criteria, and we say plainly that a formal attestation is still on our roadmap. When Juris gives a lawyer an estimate of a case's strength, it shows the reasons, calls it an estimate, never a prediction, and gives no number when the file is too thin.

You do not have to take our word for any of it. Every answer is written down on our ethics, privacy, cookies and security pages.

Whichever tools you choose, ask the five questions. The ones worth trusting will be glad you did.

Ready to add AI to your business?

Book a free 30-minute consultation: honest advice, no sales pressure.

Book a free call →

Tags

choosing softwaredata privacyAI toolsvendor questionstrust