Joonko’s former founder and CEO, Ilit Raz, has pleaded guilty to federal securities fraud. She raised approximately $27 million from investors by fabricating client references, overstating revenue, and exaggerating the capabilities of her AI technology; when investors requested supporting documentation, she submitted falsified bank statements and fake purchase orders. The case entered the judicial confirmation phase in October 2026, and if convicted, Raz could face up to 20 years in prison.
The report notes that as financing activity among AI startups surges, investors are increasingly scrutinizing the accuracy of disclosed information. According to Cornerstone Research, 15 AI-related securities class-action lawsuits were filed in the first half of 2026 alone—nearly matching the total for all of 2025 (16 cases). The legal crux is this: false statements regarding key metrics such as customer count, annual recurring revenue (ARR), and product capabilities may constitute securities fraud if they are “materially” misleading, rather than mere protected commercial “puffery.”
The article specifically warns against three common gray areas: labeling non-binding letters of intent (LOIs) or free pilot programs as “signed corporate contracts”; counting trial users as paying customers; and presenting human-assisted workflows as fully automated AI systems. It also emphasizes that D&O insurance is not a shield against fraud—standard policies typically include fraud exclusions, and Raz had already waived any claims under the company’s D&O policy in a court-approved settlement reached with Joonko in June 2025.
Author Kyle Jeziorski (Founder Shield) further advises founders to review three key provisions in their insurance policies: the prerequisites for triggering the fraud exclusion (whether a final judgment is required or whether mere allegations suffice), the conditions under which the advance‑payment mechanism for defense costs will terminate, and the insurer’s right to recover any legal fees it has advanced. The report concludes by recommending that founders establish internal guidelines for defining metrics, ensuring that all externally disclosed data can be traced back to original supporting documents such as contracts, invoices, and accounting records, while strictly distinguishing between projections and actuals, ARR and cash collected, and projects that have been contracted but not yet launched from true revenue.