LinkedIn's 2025 AI features are impressive: AI-Assisted Search eliminates Boolean strings, and AI-written InMails show a 40% increase in acceptance rates.

But here's the problem: every recruiter on LinkedIn now has access to the same AI tools. If everyone uses AI-generated messages without customization, candidates will start receiving dozens of nearly-identical InMails that all sound like they were written by the same robot.

Because they were.

Here's how to use LinkedIn's AI features as a starting point, then customize them so candidates don't immediately recognize you're using the same templates as 100 other recruiters.

The AI InMail Problem

LinkedIn's AI writes InMails by analyzing a candidate's profile and generating a message that highlights relevant skills, experience, and why they're a good fit for your role.

This works great... until every recruiter does it.

A software engineer with 5 years of Python experience at a Series B startup will start receiving messages that all say:

"Hi [Name], I noticed your 5 years of Python experience at [Startup]. Your background in [specific tech] would be a great fit for our team at [Company]..."

Every. Single. Time.

Candidates quickly learn to recognize AI-generated messages—and many start ignoring them the same way they ignore generic templates.

How to Customize AI-Written InMails

Use LinkedIn's AI to draft your message, then apply these customization techniques:

1. Add Something the AI Missed

LinkedIn's AI pulls from a candidate's profile, but profiles don't tell the whole story. Add details the AI couldn't know:

• Mutual connections: "I saw you worked with Sarah Martinez at TechCorp—she's one of my favorite people in the industry"
• Recent activity: "I saw your post about debugging production issues at 2am—that level of ownership is exactly what we're looking for"
• Company news: "Congrats on your company's Series B announcement last week"
• Industry events: "Are you planning to attend the ERE Summit next month? I'd love to connect there"

These details prove you actually looked at their profile beyond what the AI scanned.

2. Replace Generic Praise with Specific Observations

AI-generated InMails often include vague compliments:

• "Your impressive background..."
• "Your extensive experience..."
• "Your strong track record..."

Replace these with specific observations:

• ❌ "Your impressive background in product management..."
• ✅ "I saw you led the launch of [Specific Product]—the way you handled the rollout in 6 international markets simultaneously is exactly the kind of complex coordination this role requires"

Specificity is the antidote to generic AI messages.

3. Change the Opening Line

The AI-generated opening is usually the most formulaic part:

• "I came across your profile on LinkedIn..."
• "I noticed your experience at [Company]..."
• "Your background in [Skill] caught my attention..."

Try alternative openings:

• "Quick question: what was it like transitioning from [Company A] to [Company B]? I'm curious because..."
• "I have a weird request: would you be open to a 15-minute call about [Specific Topic] even if you're not looking for a new role?"
• "You don't know me, but we have 3 mutual connections who said I should reach out..."

Different openings break the pattern and signal "a human wrote this."

4. Include a Micro-Commitment Ask

AI-generated InMails usually end with a big ask: "Are you open to a conversation about this role?"

That's fine, but consider adding a micro-commitment first:

• "Would you be open to a quick 2-minute call to see if this is even worth a longer conversation?"
• "Can I send you the job description to see if it's interesting before we schedule a call?"
• "Would you be willing to answer one question about [Topic] via message? If your answer is [X], this role might be perfect"

Micro-commitments reduce friction and feel less like a generic recruiting pitch.

5. Acknowledge You're Not the Only One Reaching Out

Candidates know they're getting multiple InMails. Don't pretend they're not.

Consider acknowledging it:

"I'm sure you're getting a ton of recruiter messages (hazard of having a great profile). Here's why this one might actually be worth reading..."

Or:

"You've probably ignored 10 recruiter InMails this week. Fair. Here's why this role is different from what everyone else is pitching..."

This honesty breaks through the noise.

How to Customize AI-Assisted Search Results

LinkedIn's AI-Assisted Search is great for finding candidates quickly, but everyone using it will find the same "obvious" candidates.

Here's how to find people others are missing:

1. Layer Manual Filters on Top of AI Results

Use AI search to get initial results, then apply manual filters to find less-obvious candidates:

• Recent job changers: Filter for people who started their current role in the last 6 months (less likely to be actively recruited yet)
• Companies in transition: Search for people at companies that recently had layoffs, acquisitions, or leadership changes
• Geographic arbitrage: Search for candidates in lower-cost markets who might be open to remote roles at competitive pay

2. Search for Adjacent Skills, Not Exact Matches

AI search is good at finding exact matches. To find under-recruited candidates, search for adjacent skills:

• Instead of "5 years Python experience," search for "3 years Python + 2 years Java" (people transitioning between languages)
• Instead of "SaaS sales experience," search for "enterprise sales in professional services" (adjacent industry, transferable skills)
• Instead of "product manager at tech company," search for "product manager at non-tech company looking to transition" (less recruited, highly motivated)

3. Use Boolean for Niche Roles

AI search works great for common roles. For highly specific or technical positions, Boolean strings still outperform AI.

Don't abandon Boolean entirely—use it strategically for roles where precision matters more than speed.

Scripts: Customizing AI-Generated Messages

AI Draft:

"Hi Alex, I noticed your 6 years of experience in full-stack development at TechStartup. Your background in React and Node.js would be a great fit for our Senior Engineer role at GrowthCorp. Would you be open to a conversation?"

Customized Version:

"Hi Alex—I saw your post last week about refactoring TechStartup's entire authentication system in 48 hours when the security team found a vulnerability. That kind of calm-under-pressure problem-solving is rare.

We're hiring a Senior Engineer at GrowthCorp, and honestly, the role is less about React/Node.js skills (you clearly have those) and more about judgment calls when things go wrong. Based on that auth refactor story, you'd crush it.

Worth a 15-minute call to see if this is even interesting? If not, no worries—but I had to reach out after reading that post."

The customized version:

• References specific content (the post about the auth refactor)
• Highlights a unique skill (calm under pressure) rather than generic tech skills
• Uses conversational tone ("honestly," "you'd crush it")
• Offers a micro-commitment (15-minute call to see if it's interesting)
• Acknowledges the candidate might not be interested

The Bottom Line

LinkedIn's AI features are powerful tools—but they're available to every recruiter. The competitive advantage goes to recruiters who use AI for efficiency, then layer human judgment and personalization on top.

Think of AI as a research assistant, not a replacement for human outreach:

• AI finds candidates → You identify which ones are truly great fits
• AI drafts messages → You customize them with specific, personal details
• AI automates follow-ups → You audit them to ensure they're not spamming people

The recruiters who succeed in 2025-2026 won't be those who send the most AI-generated messages. They'll be those who use AI to scale their reach, then apply human creativity and personalization to stand out.

Everyone has access to the same AI tools. What separates great recruiters from mediocre ones is what they do with those tools.