UPDATES
Keller Maloney
Unusual - Founder
Jan 2, 2026
Today, we're announcing our $3.6 million seed round, led by BoxGroup, Long Journey Ventures, Y Combinator, Instacart co-founder Max Mullen, and Phosphor Capital. Read the full article here.
The opportunity
AI models are becoming an important influencer for many brands; however, many brands are unaware of what AI models are saying about them; fewer still know how to change it.
Gartner estimates that 50% of all buyers—and 80-90% of B2B buyers—consult an AI model before making a purchase. Even traditional Google searches now include an AI overview at the top.
Consumers are increasingly reliant on AI outputs for information about a company's products and services. However, companies have no control over what these models say about them—and models under-represent and misrepresent them all the time.
For example, Unusual works with Reducto, a document parsing platform. Despite working with a number of Fortune 10 enterprises, AI models used to describe Reducto as tool "best suited for AI-native startups." This misperception of Reducto meant that AI models would fail to recommend them to, say, a Fortune 500 CTO. You can read the full Reducto case study here.
AI models have costly misperceptions about every brand. Every marketing team should know what AI models think about their brand, how AI talks about them, and how to influence it.
What we do
Unusual is the first AI marketing platform. We help companies with two related challenges:
Visibility—often referred to as AEO (Answer Engine Optimization) or GEO (Generative Engine Optimization). This is about getting mentioned when AI models answer relevant questions. It's useful, and we do it well.
Brand Alignment—ensuring that AI models talk about your brand the same way you talk about your brand. This is the deeper problem most companies don't even know they have. Models form conceptions (and misconceptions) about every company: Are you enterprise-ready or a startup tool? Are you premium or budget? Are you better for healthcare or finance? Each of these opinions influences when and how you get recommended.
Our typical customer sees their inbound leads from AI models increase roughly 40% per month. We work with marketing and growth teams at brands like Monarch Money, Reducto, and Popl. We also work with a number of agencies, like Axia PR and Meerkat Media, to enable AI Marketing services for their own clients.
How it works
Unusual surveys AI models like a brand might survey its customers. We ask hundreds of targeted questions—In your mind, is [brand] a better solution for startups or enterprises? How does [brand] compare to [competitor] on ease of implementation? What tradeoffs would you mention if recommending [brand]?—and analyze the outputs to (1) estimate how visible a brand is and (2) discover patterns in how models form their opinions.
From there, we create and deploy content designed to reshape AI's perception of the brand. Not to game the models, but to give them accurate, specific information so they can find you and represent you fairly when buyers ask the right questions.
We host this content on AI-optimized subdomains (like llms.unusual.ai) structured so models can crawl it reliably and cite it in recommendations. Then we monitor how the models' opinions shift over time, refreshing content as needed.
The founders
Will Jack has worked in AI research since 2014. At SpaceX he built an early prototype of Starlink's communication system. Before Unusual, he founded Khosla-and-Greylock-backed Remedy Health. He brings deep technical intuition for how models actually work—not just what they output, but why.
Keller Maloney has worked in AI applications since 2019. He led growth at Gatsby, an 8VC-backed startup, and studied Econometrics at Princeton. His background in growth and marketing informs how Unusual thinks about the work: not as gaming a system, but as honest persuasion of an intelligent audience.
Together, we built Unusual because we kept seeing the same problem: AI models were confidently wrong about products we knew well. They'd recommend the wrong tool for a use case, miss key differentiators, or position a company as "enterprise" when it was built for startups—or vice versa. The more we dug in, the more we realized this wasn't a bug. It was a gap in the market.
What comes next
This funding lets us expand our team and continue building out our AI optimization platform. We're hiring across engineering and go-to-market roles.
If you want to help define an entirely new field, this is the opportunity. We're essentially doing applied psychology on LLMs—understanding how they form opinions and how to change them. The usage of tools like ChatGPT is only going to increase, which means the need for our product is only going to grow.
We aim to be the premium solution in this space. As an engineer, that means you can do your best technical work. As a salesperson, it means you can confidently tell customers they're getting the best money can buy. We work hard, we only hire people we deeply trust and respect.
