I read a security writeup this week about attackers stuffing fake support numbers and phishing links into forums, PDFs, and review pages, all tuned so that ChatGPT and Google's AI Overview repeat them as the "official" answer. They're not gaming search rankings anymore. They're gaming what the model tells you.
What caught my attention as a maker wasn't the scam itself, it's the gap it exposes. Nobody really owns "what does AI say about my brand right now, and is it correct?" Takedowns don't scale against automated campaigns, and the big vendors are calling it out of scope. That smells like a wedge for a small, focused tool: monitor how LLMs describe a company, flag poisoned answers, and give the brand something actionable.
I build SoloVault to help founders spot exactly these kinds of early signals and turn them into buildable directions.
https://solovault.info/