Shopify chief executive Tobi Lutke has cautioned employees against dumping rushed, low-quality artificial intelligence output on colleagues, describing the practice as lobbing a “slop grenade” into someone else’s workload. The comments, circulated internally and later reported by technology outlets, come as the Ottawa-based e-commerce giant continues to push aggressive AI adoption across its workforce.
What Lutke Said
According to accounts of the internal messaging, Lutke told staff that using AI tools to quickly generate drafts, code, or reports without proper review does not actually save time company-wide. Instead, he said, it merely transfers the work of fixing mistakes, filling gaps, or rewriting content onto whoever receives the output next — a manager, editor, or teammate who now has to clean up a mess that looks polished on the surface but is unreliable underneath.
“Slop grenades” — work products generated quickly with AI but not verified, refined, or fact-checked before being handed off — were singled out as a growing problem inside the company, according to reports of the memo.
The framing echoes a term gaining traction across corporate America: “workslop,” used to describe AI-generated material that appears complete and professional at first glance but requires substantial rework once scrutinized. Researchers and workplace analysts have flagged the phenomenon as a hidden cost of rapid AI rollout, one that can erode the very productivity gains companies are chasing.
Shopify’s AI-First Culture
The remarks are notable because Shopify has positioned itself as one of the most AI-forward large employers in North America. Lutke previously told staff that demonstrating AI usage would become a baseline expectation for performance reviews, and that teams should be able to explain why a task could not be accomplished with AI assistance before requesting additional headcount or resources.
That mandate has pushed employees across departments — from engineering to marketing to customer support — to lean heavily on generative tools for drafting code, writing copy, and summarizing data. Lutke’s latest comments suggest the company is now grappling with the downside of that push: workers using AI to move fast without ensuring the output is actually correct or useful.
Why It Matters Beyond Shopify
The tension Lutke described is not unique to his company. As generative AI tools become standard fixtures in offices, managers across industries have reported a rise in submissions that look complete but contain factual errors, logical gaps, or generic phrasing that requires significant editing. A widely cited concept in workplace research describes this as an invisible tax: the time saved by the person generating content is often outweighed by the time spent by others verifying or correcting it.
For companies racing to show productivity gains from AI investment, that dynamic complicates the narrative. Executives touting efficiency wins may be underestimating the downstream cost absorbed by reviewers, editors, and managers who quietly catch and fix AI-generated mistakes before they reach customers or the public.
Industry Context
Concerns about the reliability of AI-assisted output have surfaced elsewhere in the tech sector as well. Anthropic’s chief executive recently warned about the broader risks AI systems pose to internet infrastructure, underscoring that concerns about unchecked AI use extend beyond individual workplace productivity into questions of security and trust.
Shopify has not disclosed specific metrics on how much time is lost to reworking AI-generated material, and the company has not issued a formal public statement beyond the internal remarks attributed to Lutke. It remains unclear whether the company plans to introduce new guidelines, training, or review processes to address the issue, or whether it will rely on cultural messaging from leadership to curb the behavior.
What Comes Next
Lutke’s comments arrive at a moment when many organizations are still calibrating how much autonomy to give employees using AI tools versus how much oversight is required to maintain quality. Some companies have begun experimenting with mandatory AI-output review checklists or requiring employees to disclose when a document was substantially AI-generated.
- Shopify continues to require staff to justify why AI cannot handle a task before requesting more resources.
- Leadership is signaling that unreviewed AI output creates hidden costs for teams.
- The broader tech industry is still defining best practices for balancing AI speed with accuracy.
For now, Lutke’s blunt framing — comparing careless AI use to tossing a grenade into a colleague’s inbox — is likely to resonate with workers elsewhere who have experienced the same frustration: cleaning up after a shortcut that wasn’t actually a shortcut at all.
