Open your inbox. Read the last twenty cold emails. Tell me they don't all sound like the same person wrote them.
That's the arms race. Everyone got access to the same AI tools at roughly the same moment. GPT-4 made it trivially easy to generate "personalized" cold emails. A wave of tools emerged that scrape LinkedIn, reference your last three posts, mention the funding round, name-drop the podcast you were on, and end with a "noticed you're hiring for X, which tells me you're probably struggling with Y" observation that feels less like insight and more like a template with variables.
The result is inboxes flooded with outreach that is technically correct and emotionally identical. Every email checks every box of what a good cold email used to look like. Every email is suspiciously, uniformly perfect. Every email is ignored.
We've created a world where the more polished your outreach looks, the more it screams "I used AI to write this, and I'm sending a version of this to ten thousand other people today." Polish — the thing sales teams optimized for over the last decade — has become the tell. The smoother the email reads, the more obviously synthetic it feels.
This is the central irony. The technology that promised to scale relevance instead scaled sameness. The companies that bought into the automation story thinking it would give them an edge are running the same plays as their competitors, with the same prompts, against the same tired buyers. There is no edge in being indistinguishable.
I've argued elsewhere that the trust math has collapsed underneath all of this — and that the only way out runs through manufacturing warm at scale rather than spinning the volume dial harder. The arms race ends the way arms races usually end: a stalemate where everybody spent more and nobody won.
Related Essays
The Trust Collapse in B2B Outbound
AI doesn't write bad cold emails. It writes good ones. And good cold emails are now suspicious — that's the uncanny valley of sales, and it's killing response rates.
Manufacturing Warm at Scale
Stop trying to make cold outreach feel warm. Start engineering touchpoints before the ask, so the email is the last step of a relationship — not the first.
The Uncomfortable Truth About Outbound
Most sales teams want a tool that fixes outbound. The next AI feature, the next intent provider, the next personalization engine. It won't. The game changed underneath them.
Key takeaways
- Same AI tools, same prompts, same outputs — outbound flattened into a uniform style buyers now recognize at a glance.
- "Personalization" in 2025 is a template with variables, not a signal of attention.
- Polish is the tell. The more your email looks like a great cold email, the more it reads as automated.
FAQ
Why isn't AI making outbound better?
Because every team got the same tools at the same time. The output converged. What used to be a quality signal — a well-researched, crisply written email — is now the exact pattern recipients filter out.
Is personalization dead?
Real personalization isn't dead. Templated personalization is. Mentioning a podcast or a funding round in paragraph one no longer reads as effort — it reads as a variable injected into a sequence.