Before You Hit Enter: 5 AI Usage Tips to Protect Your Data

Generative AI tools are exciting, powerful, and everywhere. They’re also incredibly easy to use in a way that puts your data at risk.

That’s why we spoke with AI expert Brett Paulson, CEO of Vidrovr, to learn how professionals can take advantage of AI safely and securely. Here are his five biggest tips.

1. Read the Fine Print Before You Share Anything

Brett is in the business of AI security: his company provides AI to the US Department of Defense. But like many people, Brett shared, "I had never read the data disclosures on product websites until LLMs were widely introduced two-and-a-half years ago."

While Brett made an early shift, most people still click through terms of service without a second thought. But when it comes to AI, that quick scroll could cost you. Many platforms use submitted data to train their models, which means anything you enter could be reused or even inadvertently exposed in someone else's query.

Tip: Before using any AI tool, do a quick search: "Does [AI tool name] use my data to train its models?" If they do, give the terms of service a very close read. This simple habit could prevent serious data privacy issues down the line.

2. Don’t Mistake Convenience for Security

"Nothing is free," Brett told us. And he’s right. Free tools are enticing, but they often come with a hidden cost: your data.

"Several free versions (including ChatGPT) offer the option to protect your data," Brett noted, but you have to opt in. In general, paid versions of popular AI tools tend to be safer. That’s because you're the customer, not the product.

Tip: If you’re serious about protecting proprietary information, choose paid plans with explicit data protection policies. And make sure data-sharing settings are disabled by default.

3. Get Hands-on with Your AI Settings

"Most companies use Microsoft and Google's productivity tools which include AI services Copilot and Duet AI," Brett said. "You have to take configuring these tools as seriously as setting up your IT security posture."

Default settings may not be sufficient. Improper configurations can result in sensitive data being surfaced where it shouldn’t be. In high-stakes industries, that can be catastrophic.

Tip: Assign an internal AI lead (or partner with your IT team) to review and adjust settings for every AI feature your company uses. Treat it like you would any enterprise software rollout.

4. Make Sure Your Whole Team Understands AI Safety

Brett’s team at Vidrovr is incredibly sensitive to the Department of Defense’s strict security needs. But even if you don’t work with the federal government, chances are your work still impacts a larger ecosystem of partners and clients that expect you to keep their data safe.

Whether you're using AI to summarize meeting notes or training actual models, those activities carry risk for everyone in your orbit. It’s on you to set the standard for responsible AI use.

Tip: Establish a short list of AI guidelines within your organization. For example:

  • Never paste confidential docs into AI tools.

  • Don’t use actual client names or financial figures.

  • Outline red-flag use cases.

Clear rules make it easier to reap the benefits of AI without putting sensitive data at risk.

5. Stay Proactive, Not Paranoid

Even with the best data privacy practices, the reality is that data breaches are incredibly common – and they’ve spiked in recent years. "Overall,” Brett said, “most of your personally identifiable information (social security numbers, birthdays, etc.) has already been stolen and is located in bad actors' databases.”

But that’s not a reason to throw caution to the wind. Brett personally resets his passwords regularly, uses multi-factor authentication, and deploys credit lock services. More importantly, he recognizes that the stakes are higher when proprietary business data enters the mix.

Tip: Refresh your personal and company-wide data hygiene by…

  • Rotating passwords every 90 days.

  • Requiring multi-factor authentication on all work accounts.

  • Creating a shared doc of approved AI tools.

By building great data habits, you’ll be able to use AI more safely and effectively. 

Own Your AI Footprint

AI safety is no joking matter, Brett emphasized. "Learning the intricacies of these services and deploying them safely and to your company's advantage is not a simple task, and should not be taken lightly."

The bottom line: Serious professionals have a lot at stake. But with a few smart habits, you can use AI without compromising what matters most.

One low-stakes way to get started? Make sure you choose a workspace that puts privacy and security first. That’s Firmspace. Book a tour today to see the difference for yourself.

Darby Gerga