I learned something in the hospital (why I was in the hospital is a story for another day).

If you have ever been admitted overnight, you know that sleep is a challenge. Between the nurses, doctors, and others coming into your room every half hour, you have machines that beep and whine when they are low on fluids, or you roll over wrong and disconnect something.

One thing I learned long ago is to be prepared. I always bring a case with my tech, battery backup, and charging cables. It comes in handy for making communication easier and keeping your sanity.

One night, I was bored, so I started having conversations with Perplexity, which has a voice option for both the mobile and desktop versions.

I started by asking questions and getting answers. I wondered if it could save the answers as notes, and no, it can't access all apps, but it could save conversations as reminders on my phone. That opened up both a world of wonder and Pandora's Box in my mind.

What Is?

One of the reasons I started with Perlexity in the first place was that it was not a one-sided LLM but rather a skin for multiple platforms that searched across various perspectives and sources to compile answers. Phones make great translators of language, and I do not speak fluent medical jargon.

In today's complex and multifaceted medical system, you are faced with terms from multiple providers operating in silos. It's up to us to manage the context, stay informed, and be an active, curious consumer.

Although you may have all your medical information contained in a system like Epic's MyChart, that does not mean that all providers are aware of what others may be doing. Each issue and most organs in your body have different specialists. Each has tests, and you need to understand what those results mean.

Just looking at a blood test (CBC or a Comprehensive Metabolic Panel) can have 10 or more results that all interact like a map to your health. We know red blood cells and white blood cells, but what is an Anion Gap, and what does it mean in relation to Albumin and your BUN and Creatinine? Mind numbing, right? But it's essential.

You can easily ask Perplexity, “What does this mean, and what is the cause of it being out of the range of normal?” It's up to us to educate ourselves and ask educated questions of the people in charge of deciding treatments and more.

AI & Privacy

Thanks to HIPAA, systems like Epic and MyChart will protect your information and privacy. But that all goes out the window when we start to incorporate AI into our lives. The questions we ask and the answers we get may or may not be available to people who can both use that to help you, or use it against you in the future.

I know a few people who have created personal LLMs and uploaded their medical records and information to them, thinking they are protected by a privacy policy, but is that true? Laws and security standards usually call this kind of data “personally identifiable information (PII),” and if the data is health-related and tied to you (medical records, diagnoses, insurance numbers), it is “protected health information (PHI)” under HIPAA.

I asked Perplexity…

“Perplexity, do you store my questions and answers?”

Many AI companies monetize user data by analyzing questions and interactions to improve models, target ads, or sell insights. Companies like OpenAI have faced scrutiny over how user data may be used to refine their models and for potential monetization. Meta, with its LLaMA models, also raises privacy concerns, given Meta’s track record with user data. Other companies that have been called out for data practices include major tech giants like Amazon and some Chinese AI firms. In contrast, Perplexity emphasizes transparency and does not store your information.

“Perplexity, Should I be concerned about AI and privacy: data collection, storage, sharing, future misuse, need for regulations and transparency?”

First, data collection: AI systems often gather data from multiple sources—social media, browsing behavior, even personal devices—creating a very detailed profile of an individual.

Second, data storage: how and where this data is stored is crucial. Improper storage or weak security measures can leave data vulnerable to breaches.

Third, data sharing: companies may share or sell data to third parties, often without users fully understanding the implications. This can raise concerns about who ultimately has access.

Fourth, future misuse: AI can predict behavior, influence decisions, or even manipulate consumers. Without clear oversight, this can lead to ethical issues and potential manipulation.

Ultimately, the push for stronger regulations, greater transparency, and more user control over their data is key to addressing these concerns.

“Perplexity, although you don't save my data, are the LLMs you access tracking back to me personally?”

That's an important question. Generally, privacy-conscious assistants like me are designed to avoid storing personal data or creating profiles linked to individuals. While the questions you ask help generate responses in real time, they’re not tied back to a persistent profile. This approach helps protect your privacy and keeps the focus on your current interaction without building long-term data profiles.

Closing Thought

You should be your own best advocate for your medical care and treatments. You should also be aware of and be a good steward of the data you provide to AI engines.

If you do what I have been doing, inputting questions directly into any AI engine, you may want to be careful with the questions you ask or the data you provide. Just because they have a privacy policy does not mean you are 100% protected and safe.

If an LLM provider collects personal information and their privacy policy allows it to share or sell that data, it could have long-term implications. Your data might end up in the hands of third parties, such as advertisers, data brokers, or other companies that use it for profiling, targeted marketing, or even predictive analytics.

In the future, this could mean more personalized ads, but also potential vulnerabilities—like data being used in ways you didn't expect or influencing decisions about you. It highlights why transparency, user control, and strong regulations are so important when it comes to AI and privacy.

Your health and your data are both personal and in your control. Make sure you are asking the right questions to manage and protect both!

______________________________

Comment below and share your thoughts, ideas, or questions about business-to-business sales and marketing today! Do you have a sales or marketing communications strategy that works for you? What tips or techniques can you share that work for you and your business?

To learn more about this and other topics on B2b Sales & Marketing, visit our podcast website at The Bacon Podcast.