A paper posted to arXiv examines how workplace ambiguity—unstated expectations, evaluations, or procedures—acts as a structural barrier for autistic and intellectually or developmentally disabled (IDD) workers. The authors interviewed 11 caregivers and ran an AI co-design activity to explore technological solutions. The interviews surfaced five themes of ambiguity, which the researchers synthesized into a framework with three intersecting dimensions: evaluative, procedural, and relational.
The caregivers proposed three forms of technology support: ambient real-time redirection, feedback processing, and physiological guidance. These tools are designed to make hidden expectations perceptible and actionable, helping workers navigate situations where they would otherwise have to guess. However, the authors note a core tension: technologies that make the workplace legible to the worker can also make the worker hyper-legible to employers through surveillance.
The researchers argue that ambiguity-aware workplaces should make expectations explicit before they carry consequences, and that assistive technologies must remain accountable to the worker. The study is based on caregiver perspectives rather than direct interviews with workers, which the authors acknowledge as a limitation of the current work.