Shopify CEO Tobi Lutke has a term for a problem he says is showing up inside his own company: the slop grenade. Speaking on the podcast The Knowledge Project on Tuesday, Lutke described emails and code that employees generate with AI, barely review, and send off to colleagues anyway. "We call those 'slop grenades' that people toss at each other," he said. "That's definitely a bad thing."
The mechanism, as Lutke described it, is simple enough to recognize in any inbox. AI can produce long, polished-looking output almost instantly, and it's tempting to use that speed to skip the harder work of actually synthesizing a point. The result looks legitimate on the surface but hasn't been checked by the person sending it. "You don't really read it, and now it has to be reviewed by your colleagues, and they are like, 'this doesn't look right,'" Lutke said, describing the burden that then falls on whoever receives it.
The comment sits in some tension with Lutke's own prior stance on AI at Shopify. Last year, in a memo posted to X, he told employees that using AI was "a baseline expectation" at the company and that before asking for more headcount or resources, they should be able to show they "cannot get what you want done using AI." The slop grenade complaint isn't a reversal of that position so much as its other half: Lutke still wants AI used constantly, he just doesn't want its raw output mistaken for finished work.

