Ollie, a San Diego startup building an AI assistant for families that works entirely through text messages, said Tuesday it has raised $7.5 million in seed funding led by Khosla Ventures. AI House, formerly AI2 Incubator, also participated. The company declined to share its valuation.
Unlike the app-heavy landscape of modern tech, Ollie operates through SMS, iMessage, and MMS across both iOS and Android. Families create group chats where they can ask the AI to plan dinners, order groceries, schedule doctor appointments, or even pay bills. The assistant plugs into calendars and email, handling coordination that typically requires juggling multiple apps and browser tabs.
The funding comes at a moment when consumer AI assistants are attracting serious capital. Instinct, another player in the space, raised money at a $2.5 billion valuation in late August, underscoring investor appetite for tools that promise to simplify daily life through artificial intelligence. But Ollie is staking out different territory: it pitches itself as privacy-first in a market dominated by platforms that monetize user behavior.
"Trust and privacy are absolutely imperative," CEO Bill Lennon said in an interview. "That's why our business model is a subscription."
Rather than offering a free service supported by targeted ads or data sales, Ollie charges users directly. Plans range from a limited free tier to $100 per month for heavy users, with message caps that define how often families can ping the assistant. The company says it will never share or sell user data, and it doesn't train AI models on the personal information flowing through family chats. To avoid storing sensitive login credentials, Ollie uses remote browser sessions where users authenticate directly with services like Gmail or Google Calendar through OAuth 2.0 protocols.
The startup recently achieved SOC 2 compliance, a security certification more common in enterprise software than consumer apps. For a family tool handling grocery orders and bill payments, the designation signals an unusual level of attention to data protection.

Lennon brings experience from the intersection of consumer tech and financial services. He previously founded Groundwork, a youth-sports payments company that Snap! Mobile acquired in 2021. His co-founder and chief product and technology officer, Rushabh Doshi, worked at YouTube, Google TV, Snap, and the personal finance app Digit before joining Ollie. Christy Shannon rounds out the founding team as chief marketing officer.
The company got its start in 2023 as an AI shopping assistant while incubated at the Allen Institute for AI, the Seattle research organization backed by Microsoft co-founder Paul Allen. Since then, it has broadened into general household management. A Washington Post report from August 2025 cited more than 50,000 families using the service; current figures were not disclosed.
Ollie faces competition from a growing cluster of startups trying to crack the code on family coordination and personal assistance through AI. Poke, which Cognition acquired, operated in similar territory. So do Fambot, Ohai, Folk, Saner.ai, and Tomo. Work-focused assistants like Town, Lindy, and Reclaim.ai represent adjacent approaches to managing life's logistics. Most rely on apps or web interfaces; Ollie's bet is that text messages—ubiquitous, familiar, and already used for family coordination—provide a simpler entry point.
Vinod Khosla's venture capital firm invested $5 million in Ollie, as reported by the Washington Post. AI House is operating from an $80 million third fund following its rebrand from AI2 Incubator in mid-2026.
The company operates under the legal name Confabulation Corporation and is based in San Diego's University City neighborhood. LinkedIn pegs its headcount in the single digits, typical for a seed-stage startup still figuring out product-market fit. Whether enough families will pay for a text-based assistant, and whether Ollie can scale its privacy promises as it grows, remains to be seen. But in a market where consumer AI often comes with hidden costs, the startup is betting that transparency itself might be worth something.
