The Future of Work is Performative, and AI Can’t Fake It

his summer was rough on the AI, huh?

First, Anthropic quietly shut down Claude Explains, which was supposed to be this human-edited AI blog thing. Turns out nobody wanted to read it. Then Ramp put out this squishy announcement about how corporations might be, you know, reconsidering their AI spending sprees. They mention the Klarna thing (the AI customer support mess) and basically admit that maybe people actually want to talk to humans when their credit card gets declined. And right at the end of June, LinkedIn’s CEO drops an admission that basically no one is using their AI writing assistant. Nobody wants a bot writing their professional posts, apparently.

Cruel summer indeed.

Your Feed Is Your Resume Now

Why would LinkedIn publicly admit its AI feature is flopping? I think they are being clever here. They are positioning themselves as a serious social network while taking shots at everyone else. LinkedIn has figured something out that’s kind of brilliant in an evil product strategy way. Their value isn’t necessarily in that having a great LinkedIn presence helps your career, but in not having one actively hurts it. 

And nowhere is that fear more potent than when it comes to staying employed.

The CEO’s quote basically spells it out: “If you’re getting called out on X or TikTok, that’s one thing. But when you’re getting called out on LinkedIn, it really impacts your ability to create economic opportunity for yourself.”

Translation: screw up on LinkedIn and you can kiss your job goodbye. Maybe your house too. This is where things get interesting, though. LinkedIn isn’t backing away from AI because the writing assistant isn’t working. They’re doubling down on something bigger.

Everyone’s an Influencer (Whether They Want to Be or Not)

LinkedIn has decided that old-school resumes are basically dead. The feed is the new resume. Your posts, your engagement, your whole public persona on the platform, that’s what matters now.

Which means everyone needs to be a thought leader. Everyone needs to be posting. Everyone needs to be constantly building their brand, because that’s how you secure jobs now. Except here’s the problem: when everyone starts using AI to generate their thought leadership posts, nobody stands out. You get this weird homogenization where every post sounds the same because they’re all coming from the same models trained on the same data. And then it gets worse. It’s almost the inverse of erotic AI chat platforms, where training on real conversations actually makes the experience richer and more believable. On LinkedIn, that feedback loop kills personality. As AI learns from AI-generated content (the snake eating its tail, as it were), the whole thing just becomes this recursive loop of increasingly generic business speak no one wants to read.

The Part Where It Gets Absurd

LinkedIn’s CEO almost had me. I was ready to give him credit for recognizing a use case where AI doesn’t work. Then he goes and says this: “Every time, before I send him an email, I hit the Copilot button to make sure that I sound Satya-smart.”

He’s talking about emailing his boss, Microsoft’s CEO. And he’s admitting he uses AI to make himself sound smarter. This cannot be real. He has to be selling Copilot, right? (I mean, Microsoft owns LinkedIn, so yeah, obviously he’s selling Copilot.) But the fact that TechCrunch printed it with a straight face is kind of amazing.

At some point, we have to acknowledge our own role here. Some of us job-hopped aggressively. We turned career advancement into a game of musical chairs. Classic product mistake. Every time you make something easier, you make it more vulnerable to exploitation. LinkedIn made it easy to build a professional network, and we exploited that by turning it into a game we had to play or risk being left behind. Now we’re all stuck performing thought leadership for an algorithm, hoping it surfaces our posts to the right hiring managers, while simultaneously rejecting AI tools that would help us perform that thought leadership. The irony is pretty thick.

Where This Goes Next

Corporate leadership keeps saying “AI is a tool, not a replacement.” But they’ve been saying it while actively replacing people with AI. They expect the remaining humans to use AI as a tool or for fun, while they’re busy using it as a replacement. Maybe (and I’m being cautiously optimistic here) this LinkedIn moment signals a shift. Maybe as more people reject AI-generated content in contexts where authenticity matters, leadership will start listening to their own advice.

Or maybe we’ll all just keep posting LinkedIn hot takes about synergy and disruption until the heat death of the universe. Time will tell. Here’s hoping it’s the former.