LinkedIn’s search bar has been broken for years. Looking for an investor with FDA experience in healthcare? Better know their exact job title. Trying to locate an NYC startup founder you vaguely remember meeting? Impossible unless you can construct the perfect filter combination.
The process has been brittle, keyword-dependent, and often futile. But that has just ended. This week, LinkedIn launched AI-powered people search, a tool that reimagines how professionals discover each other on the world’s largest professional network.
The feature marks the most significant integration of AI into LinkedIn’s core product since the company began its two-year quest to infuse AI across its platform. But unlike previous AI additions, this one targets LinkedIn’s most valuable real estate, the search bar used by hundreds of millions of professionals every month.
The old LinkedIn search experience was unforgiving. You typed words, clicked filters, and hoped for the best. Users had to know exact titles, company names, or credentials to find anyone. Miss the right combination, and the person remained undiscovered.
“With lexical search, you have to know the exact title of the person, or you need to wrestle with filters to find the right person, maybe. If you didn’t know the right combination, the right person remained undiscovered. The new AI-powered people search is designed to be the fastest path to the person who can help you the most,” said Rohan Rajiv, Senior Director of Product Management, LinkedIn.
Now users can type conversational queries. “Find me investors in the healthcare sector with FDA experience.” “People who co-founded a productivity company and are based in NYC.” “Who in my network can help me understand wireless networks?” The AI understands intent, not just keywords. It translates natural language into structured matches across LinkedIn’s 900+ million profiles.
Instead of a traditional “Search” prompt, Premium users see “I’m looking for…”—a small shift in language that signals LinkedIn’s move from keyword matching to intent-driven discovery.
Why Now?
The timing is strategic. Over the past 18 months, users have increasingly migrated to ChatGPT and Perplexity for professional discovery. AI agents and browser extensions scrape LinkedIn data to answer questions LinkedIn’s native search couldn’t handle.
Reddit locked down its data and demanded licensing agreements. Google, Bing, and DuckDuckGo rushed AI search features to market. LinkedIn watched users leave and data scrapers proliferated. The company had to act.
“I think we are still early in this age of browsers and how they are working on behalf of people. I think over time, we will have a more sturdy policy around browsers,” said Rohan Rajiv, LinkedIn.
For now, LinkedIn maintains an open approach. But the subtext is clear, use LinkedIn’s native AI search instead of third-party tools extracting your data.
LinkedIn’s approach balances AI ambition with practical accuracy. The system doesn’t rely on pure generative AI, it combines natural language understanding with LinkedIn’s structured data. When you search for “people who co-founded a YC startup,” the AI maps your conversational intent to structured profile attributes. Founder status, company participation, Y Combinator involvement.
The system ranks results based on multiple signals. How closely they match your search, how connected you are to them, and how likely they are to respond. This proximity ranking matters enormously for a professional network, you’re more likely to reach out to a second-degree connection than a stranger.
Early testing revealed the system’s power and limitations. Searches for “people who co-founded a YC startup” return different results than “Y Combinator.” Queries for “voice AI startup” sometimes surface profiles with “Top Voice” badges rather than actual voice AI founders. LinkedIn acknowledges these gaps and says it’s improving query comprehension, acronym handling, and entity disambiguation.
Rollout and Strategy
Startups have already recognized the opportunity. Happenstance.ai, Superposition.ai, and Clado.ai all focus exclusively on AI-powered people discovery. These upstarts target a specific pain point. LinkedIn’s search is broken. By building purpose-built tools for professional discovery, they chip away at LinkedIn’s core value.
But LinkedIn has advantages. Startups can’t replicate. 900+ million verified profiles, years of relationship data, employment history, skill inference, and profile normalization. The company’s infrastructure for matching professionals is unmatched.
“This is sort of an area where I think it is going to be hard to find a substitute for the real thing because this is the worst the search has ever been,” said Rohan Rajiv, LinkedIn
AI-powered people search launches today for Premium subscribers in the U.S., expanding globally in coming months. This positioning is deliberate. Premium subscriptions directly benefit from faster, more efficient discovery. Better search means more conversions, justifying subscription fees.
LinkedIn’s job search AI, launched in May, followed the same playbook. Natural language job queries serve Premium users exclusively, then expand. Over 90% of early testers found the job search feature beneficial.
LinkedIn hasn’t stopped at ‘people search’. The company plans to extend AI across the entire professional workflow from estimating engagement, suggesting intros to automating outreach.
The vision is a conversational assistant that understands who you are, what you know, and what you need, then helps you achieve it. For 900 million professionals, that changes everything about how they find opportunity on the world’s largest professional network.
