Do recruiters know if you use auto apply? What hiring software vendors say (2026)
Recruiters often suspect automation, and their software flags some of it. Greenhouse reported on October 15, 2026 that 86% of US recruiters say at least half of applications are AI-written. Greenhouse and Ashby sell fraud detection that reads device, IP address, email and phone signals. No applicant tracking vendor we read says it labels an application as sent by a named bot.
How many applications do recruiters think are AI-written?
Greenhouse, which makes applicant tracking software, runs surveys of recruiters and job seekers. Two of its releases bear on auto-apply.
- In its Orchestrated Hiring Ecosystem release of October 15, 2026, Greenhouse says 86% of US recruiters report that at least half of applications are AI-written, and that fewer than 7% of applicants ever get an interview.[48]
- Its 2025 AI in Hiring Report, released November 19, 2025, surveyed 4,136 people in the US, UK, Ireland and Germany. In the US, 91% of recruiters said they had spotted candidate deception, and 34% said they spend up to half their week filtering spam and junk applications.[47]
These are Greenhouse's own surveys, published by a company that sells tools to recruiters. They tell you what recruiters believe, not how often a specific bot gets caught.
What does hiring software check on each application?
Two vendors describe their detection in public documentation.
What does Greenhouse Real Talent flag?
Greenhouse says Real Talent screens applications using objective digital signals to flag possible identity misrepresentation, lets a customer block known IP addresses, email addresses or domains, and ranks candidates against criteria the recruiter sets. Greenhouse says the matching does not automate hiring decisions.[49] Its October 2026 release says the product verifies identity and flags fraud, spam and misrepresentation at the top of the funnel.[48]
What does Ashby's Candidate Fraud Detection check?
Ashby announced the feature on September 16, 2025, saying inbound applications had become saturated with fraudulent candidates.[50] Its documentation lists four signal types it checks on each new application: device, IP address, email address and phone number. Recruiters can filter by statuses such as "Multiple Fraud Signals - Unreviewed" and mark candidates fraudulent or a false positive.[51]
Neither vendor says it identifies which auto-apply tool sent an application. Both describe signals, such as many applications from one device or IP address, that a cloud bot applying at volume could trip. That is an inference from their documentation, not something either company states about any named tool.
What does LinkedIn show employers when you use Easy Apply?
When you apply with LinkedIn's Easy Apply on a job connected through Apply Connect, LinkedIn sends the application data into the employer's applicant tracking system, and recruiters there can see a summary of your LinkedIn profile next to your name.[42] LinkedIn's own rules on automation are on our LinkedIn page.
How do you lower the risk before turning a bot on?
- Read what the tool says it writes for you. A recruiter who sees many near-identical cover letters in a day is looking at the pattern Greenhouse describes, whoever sent them.
- Prefer a tool that lets you approve each application. AIApply, JobCopilot and Dreamwork describe a review option; Simplify's free extension leaves submitting to you. See the tool files.
- Keep volume to roles you would take. A fraud filter that reads device and IP signals sees volume, not intent.
Fine Print did not run tests against any applicant tracking system. Everything above comes from the vendors' own pages, linked on the sources page.