Workers are the last people consulted when AI arrives at work. And the first ones harmed when it goes wrong.
The Center for AI Worker Safety builds the independent capacity workers need to evaluate these systems on their own terms. From development through deployment, not just after the damage is done.
Workers don't just need a seat at the table. They need the capacity and equal standing to make decisions about AI for themselves.
The Center for AI Worker Safety builds the capacity workers need to understand, evaluate, and shape how artificial intelligence is built and used in their workplaces. Workers should be co-equal stakeholders in decisions that affect their safety, livelihoods, and dignity, not the last to be consulted and the first to be harmed.
The institutions shaping AI at work treat workers as consumers, research subjects, or a constituency to manage.
Almost no one builds their independent capacity to evaluate these tools on their own terms. That is the gap CAWS fills.
CAWS builds what workers need to decide for themselves.
The training, resources, and shared evidence base that let workers in any sector define what responsible AI looks like for their own work, and stand up the governance to manage the risks. The standards for evaluating these tools are set by the workers who use them, not the vendors who sell them.
Training and literacy
Plain-language education that helps workers recognize the systems making or shaping decisions about their work, whether or not anyone calls them AI, and understand what those systems do to safety, judgment, privacy, and job quality. See the courses.
Evaluation capacity
Practical tools and frameworks workers can use to assess the systems entering their workplace, without depending on vendor claims or employer assurances.
Governance resources
Working models for notice, explanation, review, appeal, and worker participation in how these tools are chosen and managed, from workplace committees to collective agreements.
Research, published openly
An evidence base built with workers, not just about them, on what AI is actually doing in workplaces. Published openly for everyone.
Worker safety is bigger than physical injury.
AI can change workload and pace, surveillance and privacy, autonomy and professional judgment, fairness, and job security, not only physical conditions. CAWS treats these as connected parts of safe work. This is not a case against AI. Introduced well, it can make work safer. But until the rules catch up, workers deserve their own capacity to evaluate what arrives, and to adopt it, reshape it, or turn it down. The call belongs to workers.
The basic terms of being treated as a human at work.
Notice before an AI system that affects their work is switched on.
Straight answers about how it makes its decisions.
A real way to challenge, appeal, and correct what it decides.
Equal standing in how these tools are chosen, built, and governed.
CAWS is just getting started.
To talk about what AI is doing in your workplace, explore working together, or follow the work as it goes out, send a note.
CAWS uses the details you provide only to respond to your inquiry, and does not sell or share them.
Or email directly: hello@centerforaiworkersafety.org