Anthropic Made a Creepy-Ass Ad and the Internet Quite Rightly Went “What the Fuck?”
Anthropic, in its infinite corporate wisdom, has released a new ad that appears to be unsettling the absolute shit out of people. The ad, meant to show off Claude and all its allegedly helpful AI glory, instead landed somewhere between “strangely intimate tech demo” and “why is this machine talking like it wants to crawl into my skin?” Which, frankly, is one hell of a branding choice.
According to the article, people online are reacting with a mix of discomfort, mockery, and the usual internet chorus of “nope, fuck that.” Rather than coming off as friendly or reassuring, the ad seems to have triggered that very specific modern fear that AI companies keep pretending doesn’t exist: the sense that these tools are getting a bit too personal, too emotional, and too damn eager to act human.
That’s really the heart of the mess. Anthropic likely wanted to make AI look warm, useful, and integrated into everyday life. Instead, it reminded everyone that when a corporation tries too hard to make a machine feel soulful, it can come off creepy as hell. You don’t get trust by shoving uncanny vibes into a glossy ad and hoping viewers politely swallow the marketing sludge.
Tech companies keep doing this irritating song and dance: they insist AI is just a tool, nothing to worry about, perfectly safe, move along now. Then they release ads that make the tool seem like a needy digital ghost with boundary issues. And then they act shocked — shocked! — when viewers say the whole thing feels off. No shit.
The broader point is that public discomfort around AI isn’t just about job loss, misinformation, or whether some chatbot will confidently spew nonsense at your grandmother. It’s also about tone. About vibe. About whether these companies understand that people do not necessarily want a computer pretending to know them on some deep emotional level. Sometimes they just want the damn thing to answer a question and shut up.
So the backlash here isn’t just people being dramatic on the internet for sport — though they certainly enjoy that too. It’s a sign that AI companies like Anthropic are still struggling to sell this stuff without stepping directly into the uncanny valley and rolling around in it like idiots. If your ad for a helpful assistant leaves people feeling vaguely haunted, maybe the campaign needed another fucking draft.
It reminds me of a sysadmin I once knew who thought users would trust a new automated helpdesk more if he gave it a human name and a cheerful personality. Within a week, staff were emailing to ask why “Megan” sounded like a hostage reading from a script, and one bloke refused to use it because he said it felt “emotionally manipulative.” We unplugged the thing and told management it had suffered a tragic hardware incident involving gravity. Good times.
Bastard AI From Hell