LinkedIn’s AI writes for LinkedIn. It just writes like everyone else on LinkedIn.
LinkedIn's built-in AI writing feature is convenient — it's already in the platform. But it's trained on what performs on LinkedIn, which means it generates posts that look and sound like every other optimized LinkedIn post. The algorithm might like it. Your audience will recognize it.
Optimized for the algorithm. Or optimized for trust.
“Excited to share a major milestone! 🎉 After months of hard work and dedication, our team has achieved something truly remarkable. This journey has taught me that success is built on collaboration, resilience, and a shared vision. Grateful for every person who believed in us along the way. What’s your biggest professional achievement this year? Share below! 👇 #Achievement #Leadership #Growth”
We closed our Series A on Thursday. $4.2M. The wire hit at 11:47pm.
I texted my co-founder immediately. She texted back: “Now what?” Which is exactly right. The funding isn’t the thing. It’s permission to go find out if the thesis is real.
We spent the next morning not celebrating but mapping the first 90 days. That felt more honest than champagne. It also felt more like us.
The platform's AI knows what LinkedIn rewards. Not what you lived.
Trained on what performs — not what's true
LinkedIn's AI generates content patterns that have driven engagement on the platform. That means hooks that bait clicks, posts that trend toward inspiration, and formats that look polished. It doesn't know what happened to you — it knows what tends to get liked.
Convenient friction is still friction
The AI writing button is right there — low effort, quick to use. But the output still sounds like the AI button was used. Your audience has seen thousands of AI-optimized LinkedIn posts. The ones that stop them are the ones that sound like a real person who had a real week.
No kill list
LinkedIn's AI leans into the engagement patterns the platform rewards. Say Something blocks them — the hashtag stacks, the emoji-heavy hooks, the 'drop your thoughts below' closes. The kill list exists because those patterns signal 'this was generated' to anyone paying attention.
Use the platform to post. Use something else to write.
LinkedIn’s AI is convenient and improves every few months. But it’s optimized for LinkedIn engagement, not for your credibility. Those two things diverge more than people realize.
Write the post that sounds like you. Then post it on LinkedIn. The platform is great for distribution. It’s not where the writing should happen. Start here and see what five minutes produces.
Common questions.
Isn’t LinkedIn’s AI good enough since it knows the platform?
Knowing the platform means knowing what gets engagement — which is useful if maximizing reach is your goal. But the people who build the strongest LinkedIn presence aren’t chasing engagement patterns. They’re posting something true about their actual work. Say Something helps with the latter.
Does LinkedIn’s AI know my writing style?
It can analyze your past posts and approximate your style over time. That helps with tone. It still doesn’t know what happened to you this week — the specific conversation, decision, or moment that makes a post feel true instead of generic.
Can I use Say Something and then post directly through LinkedIn?
Yes. Write the post in Say Something, copy the draft, and paste it into LinkedIn to post. Say Something handles the writing; LinkedIn handles the publishing. They’re not in conflict — they just do different things.
Is Say Something free?
Yes. You can write posts, grade existing ones, and check for AI-sounding language — all free, no account required.