Perseus Labs
shopping_bag
Back to News
adtech May 6, 2026

I Worked in Advertising Tech. What I Saw There Led Me to Build Perseus Labs

I spent years inside advertising tech watching how consumer data gets collected, enriched, and monetized, often without people realizing it. Now AI is accelerating that same system, and families are feeling the consequences in real life: facial recognition in stores, deepfakes in schools, and children's images used to train systems they never opted into. That's why I built Perseus Labs, to give regular families practical tools to protect their privacy, both online and IRL.

We lock doors to protect our kids. Now we need to protect them from AI too.

Have you ever seen an ad and wondered if social media was listening to you?

Or read about a data breach and thought, great, who has my information now?

Or walked through a store, a park, or a city street and noticed you’re now in the background of someones TikTok (or reel)?

I think about my privacy in all of these scenarios because of my spent time in advertising tech (adtech). Once you see how the system works, it is hard to unsee. 

We live in a time where consumer data is treated like a raw material. It gets collected, enriched, shared, inferred, licensed, trained on, and monetized. Most people do not know what happens to their data, but its certainly in the terms of service they scroll past. My experience in adtech made me an expert in how data is used for advertising, but where does that leave everyone else?

From the outside, digital advertising can sound technical or boring. From the inside, it looks more like this: collecting signals about users, combining those signals, predicting behavior, and making money off the accuracy. Consumers feel it when an ad is a little too specific and when platforms know more about them than they intended to share.

What bothered me most was how little control consumers had in the process. People were expected to accept broad data collection as the price of using modern technology. If you wanted convenience, personalization, free products, or better recommendations, the price was more tracking. That became the default. And once something becomes the default, companies stop feeling pressure to justify it.

Families should not have to wait until their information is abused to start protecting themselves.

We are too used to talking about privacy as if it is only an online issue. It is not. It is online and in real life now.

Online, the same patterns that shaped adtech are now shaping AI and content platforms. Meta says it draws from publicly available and licensed information, as well as information shared on Meta Products, including posts, photos and captions, to develop and improve its AI.[1] Reddit disclosed that the FTC was asking questions about its licensing or sharing of user-generated content for AI training.[2] The same personal data that fueled advertising is now being reused to train the next wave of systems.

Your attention is monetized. Your activity is monetized. Your content is monetized. Now your data and likeness can help train AI too.

Offline, the story is not much better. Going to the store, entering a stadium, walking through public spaces, these are normal parts of life. Most people do not think of those moments as data collection events. Increasingly, they are.

In December 2023, the FTC said Rite Aid used facial recognition technology in hundreds of stores without reasonable safeguards.[3] In January 2023, the New York Attorney General scrutinized Madison Square Garden’s use of facial recognition to identify and deny entry to lawyers tied to firms suing the company.[4] Wegmans has also publicly acknowledged using facial recognition technology in some stores.[5]

And now we are seeing the downstream harm show up where families feel it most: schools. Scientific American reported in June 2024 that students were already using AI tools to create fake nude images of classmates.[6] The Associated Press reported in December 2025 that schools across the U.S. were facing a growing crisis involving students using AI-generated deepfakes to cyberbully peers with sexually explicit manipulated images.[7]

This is where I think a lot of current privacy conversations fall short. The solutions are too disjointed, too reactive, or too technical for the average person. Most people are not going to read every terms of service. Parents can’t monitor every place their photos, posts, and location data can travel.

And honestly, they should not have to.

That is the gap I keep coming back to.

We need solutions for online privacy and in-real-life privacy. We need tools that are practical enough for regular people, not just privacy experts. We need privacy products that do not assume people have unlimited time, technical expertise, or energy.

And for parents, that need is even more obvious. What happens when your child’s face is deepfaked by AI?

What happens when kids grow up in a world where their image is treated like open-source material for systems they never opted into?

These are not fringe questions anymore.

Big tech and governments have not given consumers enough reason to believe they will solve it on their own (or care to).

But all is not lost.

When Apple gave users a clearer choice around opting into tracking, most people who saw the prompt declined.[8] That told me something important. Consumers do care. They just need tools that are easy to understand and easy to use. That is why I started Perseus Labs.

For me, this journey is about rejecting the idea that invasive technology is just the price of participation. Consumers deserve better defaults, better tools, and better choices. We are building products that make privacy realistic for regular families.

No one is coming to save us, so we have to save ourselves.

If you are thinking about these issues too, take a look at what we are building. Explore SafeNest or learn more about Anti-AI Glasses to see how we are approaching privacy both online and IRL.

 

Sources

  1. Meta, “Privacy and generative AI.” https://www.facebook.com/privacy/genai
  2. CNBC, “FTC conducting inquiry into Reddit's AI data-licensing practices ahead of IPO,” March 15, 2024. https://www.cnbc.com/2024/03/15/ftc-investigating-reddit-over-ai-data-licensing-practices-ahead-of-ipo.html
  3. FTC, “Rite Aid Banned from Using AI Facial Recognition After FTC Says Retailer Deployed Technology without Reasonable Safeguards,” December 19, 2023. https://www.ftc.gov/news-events/news/press-releases/2023/12/rite-aid-banned-using-ai-facial-recognition-after-ftc-says-retailer-deployed-technology-without
  4. New York Attorney General, “Attorney General James Seeks Information from Madison Square Garden Regarding Use of Facial Recognition Technology to Deny Entry to Venues,” January 25, 2023. https://ag.ny.gov/press-release/2023/attorney-general-james-seeks-information-madison-square-garden-regarding-use
  5. Wegmans, “Wegmans Statement on Facial Recognition Technology.” https://www.wegmans.com/news-media/press-releases/wegmans-statement-on-facial-recognition-technology
  6. Scientific American, “Teens Are Spreading Deepfake Nudes of One Another. It’s No Joke,” June 10, 2024. https://www.scientificamerican.com/article/teens-are-spreading-deepfake-nudes-of-one-another-its-no-joke/
  7. Associated Press, “The rise of deepfake cyberbullying poses a growing problem for schools,” December 19, 2025. https://apnews.com/article/bf65455142a088824d3571a727d9a8c7
  8. Flurry, “iOS 14 Opt-in Rate: Behind the Numbers,” May 20, 2021. https://www.flurry.com/blog/ios14-opt-in-rate/