AI-generated 'workslop' is here. It's killing teamwork and causing a multimillion dollar productivity problem, researchers say (www.cnbc.com)

Researchers from BetterUp and Stanford coined "workslop" to describe AI-generated workplace content that looks polished but lacks substantive value—long, vague prose, incomplete decks, buggy code or oddly worded emails that shift the cognitive burden onto recipients. In a survey of 1,150 U.S. full-time workers, 40% said they’d received workslop in the last month and estimated ~15% of incoming content fits that description. Recipients spend on average 1 hour 56 minutes cleaning up or clarifying it (about a $186 “invisible tax” per person per month), which scales to roughly $9M/year for a 10,000-person organization. Workslop erodes trust and perceived competence—about half of workers view colleagues less favorably after receiving it—and helps explain why AI adoption has doubled (21%→40%) yet 95% of organizations report no measurable ROI. For the AI/ML community, workslop highlights structural risks beyond model quality: generative systems make low-effort, high-volume output trivial, so organizational process, provenance, and human-in-the-loop practices matter more than ever. Researchers recommend team norms (task-quality standards, disclosure of AI use), tooling for prompt/version logging and explainability, targeted training, and pilots to define appropriate use cases. Mitigating workslop requires design and policy fixes—better UX for provenance, evaluation metrics for usefulness (not just fluency), and leadership that aligns AI workflows with human accountability.