Networking in 2026: Use AI to Prepare, Not to Fake

Research before you reach out

The reason most cold messages get ignored isn't that they're badly written — it's that they're entirely about the sender. "I'd love to connect and pick your brain" tells the reader nothing except that you want something from them. Flip the whole approach: before you contact someone, ask AI to help you summarise their recent work, a talk they gave, a post they wrote, or the shape of what their company does, so you can reference something real and specific instead of something generic.

"I read your point about X and it changed how I think about Y" opens a door that "I'd love to pick your brain" will never open, because the first message proves you actually paid attention and the second proves you didn't. AI can do the homework fast — pulling together a quick, accurate picture of someone's public work in a few minutes instead of an hour of manual scrolling — but the genuine interest in what you find still has to be yours. If nothing in their work actually interests you, that's useful information too: you've picked the wrong person to message.

This research step also tells you what to actually ask for. A vague "can we chat sometime" is easy to ignore. "I noticed you moved from agency work into an in-house analytics role — I'm considering the same move and would love fifteen minutes on what surprised you" is specific enough that most people can say yes to it in under a minute, because it's a small, clearly bounded ask rather than an open-ended favour. The more precisely you can name the fifteen minutes you want and why, the easier it is for a genuinely busy person to say yes without feeling like they've just opened the door to an ongoing obligation.

Draft it, then make it human

Use AI to get past the blank-message paralysis — a short, specific, non-needy first draft that gives you something to react to instead of a blinking cursor. Then rewrite it, properly, in your own words, until it sounds like you on an ordinary good day rather than a template running on autopilot. This second step is not optional, and it's the one people skip when they're in a hurry.

Short and specific reliably beats long and polished. The tell of a faked, mass-sent message is that it's smooth, correctly spelled, and says nothing that only a real reader would recognise. The mark of a genuine one is the opposite — a slightly rough edge, a specific reference, a sentence that could only have been written by someone who actually looked at this one person's work. Recruiters and busy professionals read dozens of these a week; they've all developed an instinct for which category a message falls into within the first line.

A useful habit: read your draft back and ask "could I send this exact message to ten different people unchanged?" If yes, it's not personal enough yet, however well AI polished the prose. Go back and put in the one detail that only applies to this person — a specific project, a specific line from something they wrote, a specific reason their work is relevant to what you're doing right now.

Where AI genuinely helps

Lean on it for preparation, not pretence — there's a real, useful role for it at every stage of building a relationship:

• Homework — a quick, accurate brief on the person or company before you write, so your first message references something real.
• First draft — a starting point that gets you past the blank page, which you then rewrite to be personal and genuinely short.
• Rehearsal — practise out loud what you'll actually say at an event, in a call, or in an interview, including the awkward follow-up questions.
• Follow-up — a warm, specific thank-you afterward that references something actually said in the conversation, not a generic "great meeting you."

The relationship is still yours

AI can prepare you thoroughly for a conversation; it cannot have the relationship on your behalf. The trust that eventually turns a contact into a referral, an introduction, or an opportunity is built in small, real, human moments — a thoughtful reply sent at 11pm because you actually cared, showing up to something you said you would, remembering a detail from a conversation three months later. None of that is automatable, and trying to fake it with AI-generated warmth is usually more obvious than people expect.

Used to prepare, AI quietly makes shy or busy people braver and better at something they'd otherwise avoid entirely. Used to fake volume, it makes everyone who touches it forgettable, and it can damage your name faster than doing nothing at all — a bad mass-message is worse than no message. Send five real, well-prepared messages a week instead of five hundred hollow ones a day. That trade — depth over volume — is the entire difference between networking that compounds over a career and networking that gets you quietly ignored.

This matters more, not less, for people who find networking genuinely uncomfortable. If the thought of cold outreach makes you want to close the laptop, that discomfort usually comes from not knowing what to say — and thorough preparation is a real, practical fix for exactly that feeling, in a way that generic advice to "just put yourself out there" never is. Walking into a conversation having actually done the homework changes how it feels in your body, not just how it reads on the page. That's the quiet, underrated value of using AI to prepare properly: it doesn't just improve the message, it makes the whole activity less frightening to attempt in the first place.