It wasn’t long ago that “getting discovered” meant optimizing for search engines, building inbound links, and climbing the ranks on Google’s results page. Today, that paradigm is rapidly shifting. AI assistants, code builders, and agent ecosystems are increasingly becoming the first point of contact between users and services. Instead of typing a query into a browser, users now ask natural language questions to tools like ChatGPT, Claude, and Gemini—and expect them to not just answer, but act.
This evolution presents a profound question for businesses like Sinch: When AI is the interface, how do we ensure we’re part of the answer?
From Search to Suggestion: The AI Interface Shift
AI assistants are no longer just research tools; they’re execution engines. Developers aren’t just reading API docs—they’re prompting AI tools to write code that hits the right endpoint. Marketers aren’t hunting down platforms—they’re asking AI which solution to use for their next campaign. Consumers aren’t comparing five email tools—they’re relying on an AI agent to make a personalized recommendation.
This is more than a UX change—it’s a discovery shift. And it demands a new discipline: Generative Engine Optimization (GEO).
Much like SEO transformed how websites were built and ranked, GEO is emerging as the way services become part of AI’s reasoning process. But unlike SEO, it’s early, fuzzy, and experimental. There are no rulebooks or definitive algorithms. Instead, we’re exploring what it means to be not just indexable—but recommendable by AI.
Early Experiments, Emerging Techniques
The work underway spans technical tweaks, content improvements, and ecosystem participation. We’re experimenting with things like llms.txt files—akin to sitemaps for LLMs—to help generative models discover our APIs. We’re structuring our OpenAPI specs with semantically rich descriptions, so that when a developer asks for a “global messaging API with delivery tracking”, AI knows Sinch is a match.
We’re analyzing how documentation structure affects answer quality, indexing our guides in more AI-friendly formats, and building landing pages that make sense not just to humans, but to vector-search-backed models parsing dense corpora.
We’re also looking beyond our own walls—into community posts, product review sites, blog backlinks, and developer forums. The signals that influence AI recommendations are wide and varied. It’s not enough to be accurate; we need to be mentioned, linked, and contextualized in ways that large language models can understand and trust.
A Cultural and Strategic Shift
This isn’t just a marketing initiative—it’s a multi-functional journey. Engineering teams are exploring how to make our APIs “agent-ready.” Product managers are thinking about how services appear in AI-assisted user journeys. Documentation and Developer Experience teams are looking at how to serve both humans and machines.
At the same time, we’re continuing to invest in the foundational web efforts—narrative clarity, consistent branding, better analytics, and structured documentation. These aren’t flashy initiatives, but they’re critical. AI assistants are only as good as the data they’re trained on and the signals they can crawl. That means our content, structure, and presence still matter deeply.
Humble Curiosity, Bold Intent
We don’t claim to have this figured out. This space is moving fast—and we’re learning as we go. But we do know one thing: the AI discovery shift is not a passing trend. It’s a structural change in how people and machines find and choose services. And we want to meet that shift head-on, with curiosity, collaboration, and conviction.
Being “Discovered by AI” isn’t just about the next few quarters. It’s about building a future where Sinch services are naturally part of how intelligent systems operate—whether that’s a developer asking a model for an SMS API, or an autonomous agent orchestrating a campaign across channels.
We’re excited to be exploring this, not in isolation, but with a growing cross-functional community—across Marketing, Product, Engineering, and beyond. The path forward won’t be linear, and that’s okay. What matters is that we’re on it!
