Ditto, a Gen Z dating app that replaces swipe-based matching with AI-driven matchmaking, has reached 150,000 signups since launching in early 2025. Co-founded by Allen Wang and Eric Liu after they dropped out of UC Berkeley, the app is expanding across California college campuses and reports over 161,389 students joined on its website as of early August 2026 (TechCrunch; ditto.ai).

Ditto's onboarding is built around conversational AI rather than profile construction. Users sign up by texting an iMessage number with a code. An AI chatbot then collects biographical information, personality traits, interests, and dating preferences, and even accepts photos of celebrity crushes as input. The algorithm matches people based on what their hobbies and interests reveal about deeper character, rather than surface-level similarity (TechCrunch).

The delivery mechanism is deliberately throttled. Every Wednesday at 7 p.m., Ditto messages users with their match for the week, including the person, the time, and the location for a date. The app also generates collage-like graphics that introduce the matched users to each other and explain why the algorithm paired them (TechCrunch).

About 20% of matches connected through Ditto end up going out on a date (TechCrunch). The company raised a $9.2 million seed round, according to a pitch deck reviewed by Business Insider (Business Insider). Ditto was founded in 2024 and launched its product in early 2025, with initial traction concentrated at California universities (New York Post; cleveland.com). Its website currently promotes an enrollment deadline of August 12, 2026 for the next round of matches (ditto.ai).

The architecture is worth pulling apart. Ditto is not using an LLM (a large language model — the type of AI behind tools like ChatGPT) to generate better recommendations within a Tinder-style interface. It is removing the browsing surface entirely. Users never scroll through profiles or make binary yes/no decisions. The AI agent acts as a gatekeeper: it takes in unstructured conversational and visual input, infers personality dimensions, produces a match, and schedules the logistics. The swipe-based model that has dominated mobile dating since Tinder's 2012 launch is replaced by a push model where the platform delivers one curated date per week.

This is a meaningful design bet. Swiping optimizes for engagement and retention — it keeps users in the app. Ditto's weekly-cadence model pushes users off-platform and into a real-world meeting, which is the opposite of the engagement-loop logic that has governed dating app economics for over a decade. The 20% date-conversion rate is the metric that matters under this model, and it is the one investors will watch. Traditional dating apps rarely disclose match-to-date conversion, but industry estimates have historically been far below that threshold.

The iMessage-based onboarding is also a deliberate friction choice. No app download is required to start the process, which likely reduces signup friction at the top of the funnel. Whether Apple's evolving messaging policies and the eventual need for a native client will force an architecture shift is an open question.

The competitive landscape is not standing still. Tinder itself has been experimenting with social matchmaking mechanics since at least 2023, when it announced a feature letting users' friends and family recommend matches on their behalf (TechCrunch). The direction of travel across the industry is away from the pure swipe-toward-decay loop and toward curated, lower-volume, higher-intent models. Ditto's approach is the most aggressive removal of user agency in matching that has reached meaningful scale.

The constraint worth watching is whether Ditto's pool remains dense enough at the campus level for the algorithm to produce geographically viable matches week after week. A 150,000-user base concentrated across California colleges provides density in some markets and thin coverage in others. Expansion beyond the campus footprint will test whether the model works when users are not within a few miles of each other.

Ditto's current trajectory suggests the AI-matchmaking thesis has found early product-market fit with a cohort that never formed deep habits around swiping. Whether 20% date conversion scales with user growth, and whether the weekly-cadence model holds once the novelty factor recedes, are the two variables that will determine whether this is a category shift or a campus phenomenon.