Intelligence-driven insights for modern risk leaders

Instagram doesn't have groups, so the Facebook playbook from Part 1 wouldn't work. There was no gated community to get inside, no admin to convince. I had to flip the model: build the fake 14-year-old again, this time with an AI-generated face pulled from an account on the platform itself, post into the open, and let the predators find me. They did. Part 2 is what Instagram's own systems did next: a search filter that failed on a single added letter, a recommendation engine serving accounts that solicited CSAM under the banner "Suggested for you," and a WhatsApp number that took more than 250 calls and messages in ten days. The platform wasn't just failing to stop these users. It was introducing them to me. On August 6, 2026, a court in Santa Fe reached the same finding, and put it in an order.

On March 24, 2026, a Santa Fe jury ordered Meta to pay $375 million, the maximum penalty on all 75,000 violations, for what its platforms failed to disclose to New Mexico's children. A bench ruling in August added a public nuisance finding and another $567 million. I was one of the investigators whose work helped build that case. Part 1 is what I found on Facebook, and how. I bought an aged account for a few dollars, rebuilt it as a 14-year-old girl, and watched every tripwire stay silent. Facebook let the profile join teen dating groups with age ranges starting at eleven, then recommended more, including one named "Girls under 13 years old." The first solicitation arrived after a single post. That's all it took.

Risky Business Solutions featured on WIRED for our investigation exposing Gen-AI clickbait factories

Risky Business Solutions co-founder Sandeep Abraham joined WBUR’s On Point to dissect the systemic failure of social media platforms to curb predatory advertising. Discussing recent revelations that platforms profit from fraudulent ads, Sandeep argued that internal safety measures are compromised by revenue incentives, underscoring the critical need for external red teaming to expose these algorithmic vulnerabilities.

Risky Business Solutions conducted a comprehensive Red Team assessment to stress-test AI moderation safeguards against adversarial attacks. The engagement included a strategic Threat Intelligence report analyzing how bad actors leverage generative AI on the dark web for scams and disinformation.

Risky Business Solutions co-founder Sandeep Abraham provided expert analysis for a Reuters investigation exposing Meta’s $16 billion revenue from scam ads.

Risky Business Solutions co-founder Tony Eastin joined Iowa Public Radio to expose a growing threat in AI-driven disinformation. Their investigation revealed how "dark PR firms" are purchasing expired local news domains to host AI-generated content, effectively weaponizing the historical reputation of trusted journalism for automated influence operations.

Risky Business Solutions conducted a deep-dive investigation into "Dark PR" firms weaponizing generative AI to manipulate financial markets. The report exposed a network of "zombie" newspapers and fabricated journalists effectively tricking Google News algorithms, offering critical threat intelligence for organizations needing to detect synthetic influence campaigns that evade traditional safety filters.