AI gives you the tools to do a lot of things in a fraction of the time they used to require. But using AI has its privacy costs, and trying to avoid them can lead to some really bad consequences. AI also doesn’t always get things right.

The most important thing you need to know about AI is that it’s not private. As soon as you put something into AI, it’s out in the world. AI makes its living by collecting data from the internet – lots of data at lightning-fast speed and often without permission. That means they frequently gather sensitive information without your knowledge or consent, and that data can be used for purposes you didn’t agree to, raising ethical concerns. AI can also enable unchecked surveillance, and when you combine that with the vast amounts of data stored by AI systems, you could be set up for the heightened risk of unauthorized access and data breaches. That’s bad and ugly.

Protecting your privacy takes some work. Best practices include never sharing sensitive information, actively controlling data settings, and using approved AI tools. Additionally, regularly deleting past chat histories and reading privacy policies can help ensure better data security. To be anonymous, consider using tools that run locally on your device instead of cloud-based services, and utilize a secure browser with privacy features. You can also use a VPN and sign up with a disposable email address to further protect your identity. DO NOT try to use the dark web. I have some good “street smarts” in the internet world, but these streets are populated by some really bad people. You can easily stumble into a really bad place and be ruined. It’s not like a game.

You also need to remember that AI is not always correct or all that smart. AI can provide wrong answers, so it’s up to you to make sure you verify any information you get from it. A wrong answer could create a life-and-death situation, a costly business decision, or a major inconvenience – just to name a few.

But what I find really frustrating is when Copilot puts you into a loop that is not related in any way to what you’re trying to do. I understand how computers work – they’re anal – but you are supposed to be able to use plain English get your work done or get information. I got into some serious verbal tiffs with my AI; it was really ugly.

However, if you check and double-check AI and really stay on top of it, the technology can work wonders. As an example, I was able to create in just a few hours a travel itinerary that combined independent travel and a cruise. It got us down to the minute we would arrive in various locations, and in the next-to-the-last stage, it gave us choices of hotels that met specific criteria. On the downside, it included a hotel that doesn’t exist, but the beauty was that as I continued to feed it information, it kept narrowing down our options to those that were more acceptable. The lesson from this exercise is that you have to verify all the information, but the bottom line is that it saved me days of work.

We can help you peel away the bad and the ugly to get to the good stuff in AI. Call us – 973-433-6676 – or email us to see how you can keep your corporate data safe and prevent employees from leaking sensitive information. We can set up rules for how your system uses AI. Our new security software tools provide safeguards if you choose to use them.