Some 70% of Job Seekers Use AI to Research Their Future Employer
Indeed data shows some 70% of job seekers use generative AI tools to research employers, while roughly 40% of cited culture sources come from Glassdoor and Indeed, not career pages.
By Grace Kim
3 min read

What's News
- Some 70% of job seekers use Claude, ChatGPT, Perplexity and other generative AI tools to research employers, according to new Indeed data.
- Roughly 40% of sources cited in culture answers come from Glassdoor and Indeed rather than company career pages.
- Taylor Meadows calls for AI audits that ask different LLMs what it is like to work at the company.
- Meadows says departing employees’ reviews can later surface through AI, making offboarding important.
- Companies should publish employee and candidate Q&As on social media and third-party review sites.
By Grace Kim
Some 70% of job seekers now use generative AI tools such as Claude, ChatGPT and Perplexity to research employers, according to new Indeed data.
That finding puts pressure on human resources teams because job candidates can now ask chatbots what employees really think of a company. Roughly 40% of the sources cited when job seekers ask about company culture come from Glassdoor and Indeed, rather than company career pages, the data shows.
The mix matters because the tools can synthesize reviews and other information already online. A polished career page may tell one story, an employee review another, and an LLM can combine them into an answer for a candidate. Indeed’s data shows AI shaping employer perceptions from material HR teams may not own or edit.
What can employers actually control?
Taylor Meadows, principal advisor at Glassdoor, said: “AI isn’t uncovering anything that isn’t already online.” For Meadows, the issue is participation, not invention. “Your employer brand exists whether you are participating in it or not,” he said.
His advice is to feed the beast, meaning employers should engage with the questions candidates now put to the models. The recommendation treats AI-mediated research as an active channel for employer communication, not a side effect of recruiting.
How should employers run an AI audit?
Meadows recommends an AI audit as the first step. Employers should ask different large language models: “What is it like to work at X company?” The audit should surface the problems that emerge, then “reverse engineer” them. It gives HR a way to compare recruiting messages with employee evidence available to candidates.
Meadows’s sequence starts with the outside answer, then works backward to the reputation issues behind it. That reverses a conventional review process in which companies check only their own pages, because candidates can compare those pages with reviews inside the same chatbot answer.
What should companies publish?
Meadows says companies should publish Q&As that address real questions from employees and candidates on social media and third-party review sites. His recommendation puts responses beside the reviews and gives candidates direct company context, rather than leaving a career page as the final word.
His standard is candor: “Let that be the good, the bad, and the ugly.” Meadows said: “It’s going to be an opportunity for you to talk about what’s going well, but then it’s going to also be an opportunity to exercise some humility.”
Meadows’s approach puts the material outside employer-authored campaigns. The Q&As should address the same issues candidates can prompt a chatbot about, while showing where conditions differ across the company.
Why do team differences and departures matter?
Companies should also create content that shows what it is like to work in different departments, Meadows said. He notes that culture can vary by team, so a single company-wide account can leave out meaningful differences. Encouraging employees to share authentic experiences helps give candidates that detail.
Offboarding is part of the same issue. Departing employees may leave reviews that AI later surfaces, and Meadows says their final impression becomes especially important. Their offboarding experience can therefore become source material in a future candidate’s employer research.
Meadows’s conclusion leaves employers with a limited objective. “None of this is about perfection and none of this is about control,” he said. “It’s just about informing and influencing.”
What should HR do next?
The first step is Meadows’s audit: ask several LLMs what it is like to work at the company, identify the problems that surface and work backward from there. The second is to publish direct answers on social media and third-party review sites, with material that reflects differences among teams and authentic employee experiences.
The next test for HR is whether those moves can shape the employee evidence that appears in AI answers before candidates receive them. Indeed’s 70% usage rate and the roughly 40% share of culture-answer citations from Glassdoor and Indeed make employer participation a current operating requirement, not a communications option.
Source: fortune.com
More from Grace Kim
Show full bio
Market editor covering people, launches and day-to-day practice in entrepreneurship, startups and venture funding at Business Bearings.
1,719 articles