Former Google executive Justin Inman is betting that the next major battleground for brands won’t be search rankings or social feeds—but AI-generated answers.
His startup, Emberos, announced Monday it raised $1.2 million in pre-seed funding from a single angel investor. Inman, who is Emberos’ founder and chief executive officer, declined to disclose specifics.
Founded last year, the company is positioning itself for a moment when discovery increasingly happens inside AI-generated responses rather than through blue-link search results.
Emberos works across five major AI platforms—ChatGPT, Gemini, Claude, Perplexity and Grok—and plans to expand into voice assistants like Alexa. Its pricing is based on the number of agents a client deploys, spanning monitoring, predictive strategy, workflow orchestration, agentic shopping and AI governance, Inman said.
The pitch comes as brands grapple with how large language models interpret and sometimes misrepresent them, often without visibility into where those narratives originate or how they calcify over time.
“At our core, we help brands to understand and control how they’re represented in AI answers—not just from a visibility standpoint, but also recommendations across all the major models,” Inman said. “We are the infrastructure layer for what we call AI brand orchestration.”
The market for AI answer engine optimization is becoming increasingly crowded as marketers adjust to a shift in how people discover information, a pull away from traditional search and toward summarized chatbot responses. Startups including Profound and BlueFish have drawn investor interest by helping brands understand how they surface inside AI-generated answers.
Emberos has 14 employees and is using its pre-seed funding to invest in engineering while building partnerships and service capabilities, Inman told ADWEEK, adding that the company plans to raise a seed round by the end of the first quarter.
The company is working with entertainment studios, ad agencies, consumer brands—including celebrity-owned brands—and political organizations. Inman declined to name clients.
From monitoring to prediction
Rather than reporting outcomes after the fact, Inman claims the platform predicts how brands will appear in AI engines, and pushes recommended fixes directly into existing marketing workflows.
“A lot of our competitors are looking at it from a reactive standpoint—telling you what has already happened. We have a predictive capability, and then we give you what we call fix packs,” Inman said. “I want to demystify that black box and make it more approachable and automated—not just for the executive level, but also the practitioner level.”
Those “fix packs” are executed through integrations with tools like Slack, HubSpot, and Jira, routing recommendations directly to social, marketing, or agency teams, Inman said. Emberos also includes a governance layer focused on hallucinations, intellectual property concerns, and brand safety.
Mapping the AI narrative layer
At the core of Emberos’ technology is a proprietary Brand Knowledge Graph, which maps entities such as brands, products, creative assets, and narratives across the internet. That system allows the company to track how those entities evolve—and to predict future outcomes.
“Once an entity is mapped over to our Brand Knowledge Graph, we have a city map of how the internet—all the biggest influencing factors—relate to that specific entity,” Inman said. “That’s what allows us to have a predictive capability.”
Those predictions can span macro outcomes like box office performance or more granular business impacts tied to specific changes brands make. “If you make this change, we predict you’ll have X amount of lift,” Inman said. “Then once that change actually happens, we can compare what we predicted to what actually occurred.”
Unlike tools that encourage brands to flood the web with more AI slop, Emberos evaluates signals across owned, earned, and paid channels, including website FAQs and PR placements as well as influencer partnerships and paid campaigns.
“A lot of what we’re seeing in the space is primarily focused on just content and this push to publish,” Inman said. “That’s creating a lot of AI slop.”
The urgency, he argues, is already visible. According to Emberos’ data, 90% of 500 brands the company surveyed have factual errors across at least one major language model.
“That’s massive,” Inman said. Those errors can include incorrect pricing, outdated product details, or conflicting recommendations across models. Once embedded, they’re difficult to unwind.
“Once AI systems understand a narrative, that narrative hardens pretty quickly,” he said. “Our pitch is get in now and start shaping these things, and start preparing your brand for this point of discovery that has shifted upstream.”
