Stock AI Replies to Recruiters Got 12% Fewer Calls. Propose Two Times

A recruiter's message lands, the role looks plausible, and you have ten minutes between meetings to answer it. The easy move is to paste it into ChatGPT and send back whatever it gives you. Pin looked at 160,000+ candidate replies in an analysis published in October 2026, and by that data the easy move costs you. Replies with stock LLM phrasing moved toward a scheduled call 27.7% of the time. Plainly written replies got there 31.5% of the time.

The gap is real but small. The more useful part is why it shows up and what a reply that turns into a call actually contains.

What did Pin's 160,000-reply study find?

Plain replies to recruiter outreach moved toward a scheduled call 31.5% of the time, and replies with recognizable stock LLM phrasing did so 27.7% of the time. That is a 3.8-point gap, which works out to roughly 12% fewer calls in relative terms.

How the study was built:

• Sample: 160,000+ substantive candidate email replies sent between January 2025 and September 2026.
• Filter: only replies of 300 to 4,000 characters, which drops one-liners and auto-replies.
• AI-styled subset: about 2,000 replies, roughly 1.3% of the sample.
• Trend: from 2024 through Q3 2026, the share of replies with recognizable AI stock phrasing never left the low single digits.

There are two caveats. This is a correlation, not an experiment: nobody randomly assigned AI drafts to candidates. And Pin is a recruiting vendor, not a neutral academic lab. Still, the direction matches what job seekers already suspect. Greenhouse's Future of Hiring Candidate Report came out on Oct 22, 2026, and it surveyed more than 2,800 job seekers across five countries, 1,200 of them in the US. Among the US respondents, 39% had already rewritten an application so it wouldn't sound AI-generated, and another 21% had thought about it. Meanwhile 77% of US job seekers used AI somewhere in their search in the past year.

Greenhouse CPO Sharawn Tipton described both sides of the problem: "Candidates send applications into a black hole, while employers sift through AI-polished resumes that all look the same."

Why does stock AI phrasing cost you calls?

Stock phrasing costs calls because it fills the space where your specifics should be, not because recruiters catch AI and punish it. Pin's own reading of its data is that relevance to the candidate drives replies, not whether a model wrote the first draft.

Think about what the recruiter is deciding. They're choosing whether to give you a slot on a full calendar. To decide, they need three things:

1. Are you plausibly a fit for this specific role?
2. What would it take to move you?
3. When can you talk?

"I'm excited to explore how my skills and experience align with this opportunity" answers none of those. It's polite noise. A model asked to "reply professionally" will reach for exactly that kind of sentence, because it's what fits every reply to every recruiter.

The second problem is pattern matching. Recruiters spend their days around templated, AI-assisted outreach, and they've learned to skim past it. A reply in that same register looks like low intent, even when you meant every word.

The best counterweight comes from economists Emma Wiles, Zanele Munyikwa, and John Horton. In their NBER field experiment, jobseekers who got non-generative algorithmic writing help on their resumes (grammar and clarity fixes, no generated text) were hired 8% more often, at 10% higher wages. So clearer writing helps you. Generic content hurts you. The two findings fit together once you separate editing from authorship.

Should I use ChatGPT to reply to a recruiter?

Use ChatGPT to edit your reply, not to write it. Type the facts yourself (your fit, your questions, your availability), then let a model tighten the wording without adding anything new.

A workflow that keeps the help and drops the penalty:

1. Draft the content in your own words first, even as rough bullets. The model can't know which project maps to this role. Only you can.
2. Ask for edits, not a rewrite. For example: "Shorten this. Fix grammar. Do not add greetings, enthusiasm, or any sentence I didn't write."
3. Delete anything you wouldn't say out loud to the recruiter on a call.
4. Check that every sentence holds a fact: a project, a number, a question, or a time.

Don't overcorrect either. The data doesn't say you'll be blacklisted. The measured gap is 3.8 points, and it applies only to replies with recognizable stock phrases. A plain, specific reply wins. A carefully checked AI-assisted draft is not a disaster.

Half of interested candidates never propose a time

In Pin's intent data, 62.1% of candidate replies show positive engagement, but only 30.7% actually try to schedule a conversation. The other 31.4% express interest and stop there. If you want the call, put times in the first reply.

"Sounds interesting, happy to chat" hands the next step back to the recruiter. They have to write back, ask when you're free, and wait. That's an extra round trip, and conversations go cold over round trips. Pin measured the same decay from the recruiter's side: when a recruiter wrote back within an hour, candidates kept the conversation going 47.4% of the time. When the recruiter took more than three days, that fell to 35.7%. Delay costs you on both ends of the thread.

The fix is one sentence: two windows with a time zone. "I'm free Tuesday 2 to 4pm ET or Thursday before 11am ET" does more than three sentences of enthusiasm. The recruiter can book straight from your reply.

How to reply to a recruiter message so it becomes a call

A reply that turns into a call has five parts, in this order, and fits in a short paragraph or two. Each part answers a question the recruiter has to settle before booking.

1. Name the role. Thank them in one clause and repeat the title, so the thread is searchable and they know you read the message.
2. One fit line, with proof. Pick the project or result that maps most directly to what the posting asks for.
3. One or two deciding questions. Comp band, remote policy, team scope: whatever would actually change your answer.
4. Your constraint, if you have one. Notice period, confidentiality, a start-date window.
5. Two time windows, with a time zone.

The fit line is the hard part. It means reading the posting, picking the one piece of your history that matches it best, and phrasing it so a recruiter gets it at a glance. That's the same matching Refolk does for a resume: paste the posting, and I score how well you actually fit and rewrite your own resume for it. The strongest line in that tailored resume usually makes the best fit sentence for your reply, too.

Reply to recruiter LinkedIn InMail example

Here's a plain reply to an analytics engineering pitch. It sits comfortably inside the 300 to 4,000 character range Pin studied.

Hi Priya, thanks for reaching out about the Senior Analytics Engineer role.

It looks like a close fit. I own our data models and warehouse today, and I led a warehouse migration much like the one described in the posting.

Two things I'd want to know before a call: the base salary band, and whether the team is remote or hybrid.

I'm currently employed, so I'd appreciate discretion. I'm free Tuesday 2 to 4pm ET or Thursday before 11am ET. Happy to send a resume ahead of time.

Thanks, Sam

Count the facts: role, proof, two questions, a constraint, two windows, an offer of a resume. No sentence is filler.

Responding to a recruiter email when you're interested but employed

If you're interested but already employed, say so plainly and ask for discretion in one line. That describes most repliers: in Pin's data, 69% of candidates who answered recruiter outreach weren't actively job-hunting.

The hard part for employed candidates is volume, and the quiet worry that goes with it. Pin found that a candidate contacted by one recruiting team replied 14.5% of the time, but once ten or more employers had reached that same person within 90 days, the response rate dropped to 5.6%. People get swamped. You end up juggling ten threads on your phone at lunch, rewriting the same fit paragraph each time, and hoping none of it gets back to your manager.

If you are going to answer by hand, triage hard. Reply fully to the two or three roles that clear your bar. Send the rest a single polite line: "Not a fit right now, but please keep me in mind for staff-level platform roles." That decline is still a reply, and it tells them what to send you next time.

Who reads your reply: 29,178 US recruiters in Refolk's index

In Refolk's index of professional profiles, 29,178 US profiles are currently titled Technical Recruiter, Talent Acquisition Partner, or Sourcer, compared with 2,613 in the UK. Only 303 of the US recruiters mention "AI" in their headline.

The full picture side by side:

Three things follow from these numbers.

Titles hint at who's writing to you. In a small 25-profile sample, Technical Recruiter was the most common US title (15 of 25) and Talent Acquisition Partner the most common UK title (21 of 25). Treat that as directional, but it suggests UK messages more often come from in-house generalists, while US messages more often come from tech-specialist or agency recruiters. US employers in the index range from in-house teams at Snowflake, Blue Origin, and H2O.ai to staffing firms like Experis. UK employers include AstraZeneca, BT Group, and Hays.

Match your detail to the reader. A Technical Recruiter can take the stack-level detail: which systems you migrated and how much of them. A generalist TA Partner hiring across functions will move faster on the outcome: "cut our reporting refresh from overnight to hourly."

Assume a person is reading. Only about 1% of US recruiters put AI in their headline. Whatever tools they use behind the scenes, the person choosing whether to book you is reading your words, and a template reads like one.

What to cut from a recruiter reply

Cut every sentence that would be equally true of any candidate replying to any recruiter. If a line could be pasted unchanged into someone else's reply, it's taking the space your relevance needs, and Pin's data says relevance is what moves replies.

FAQ

How long should a reply to a recruiter be?

Short enough to carry five parts and nothing else: role, fit line, questions, constraint, times. That fits in a paragraph or two. Pin's study only counted replies between 300 and 4,000 characters, and what separated the better-performing replies was relevance, not length. Anything past those five parts is usually filler a recruiter will skim.

Should I reply to a recruiter if I'm not interested?

Yes, in one line. Pin found that response rates fall from 14.5% to 5.6% once a person has been contacted by ten or more employers within 90 days, so recruiters expect silence. A short decline that names what you would consider ("staff-level platform roles, remote") still counts as a reply, and it shapes the next message they send you.

Do recruiters run AI detectors on candidate replies?

Pin's data doesn't show detection. It shows outcomes: replies with recognizable stock phrasing moved toward a call 27.7% of the time versus 31.5% for plain ones. The likelier explanation is that generic wording carries less of the relevance a recruiter needs to book a call. Write something specific and the detection question stops mattering.