"We appeal as human beings to human beings: Remember your humanity, and forget the rest.” (Russell-Einstein Manifesto, 1955)

Our nation is in a state of turmoil due to the violent response to protests over the murders of people of color by U.S. police. Recent events have shone a spotlight on the brutal and racist policing behaviors of our nationwide network of local law enforcement agencies, as well as the dangers of a police force that is heavily militarized, unaccountable, and untrained or unwilling to build ties with their communities. In cities across the U.S., police have been deploying chemical weapons, vehicles, and less-than-lethal rounds, against peaceful protesters, children, bystanders, and members of the press -- especially against members of our BIPOC (Black, Indigenous and People of Color) communities. While the technologies used in the service of police brutality has, thus far, been confined to those technologies described above, there is significant risk of future misuse of robotic technologies by police forces, especially as police forces around the world continue to adopt robotic technologies.

We acknowledge that there are excellent researchers doing valuable work with local law enforcement agencies, and that there are local police departments that have not demonstrated brutality. We also acknowledge that law enforcement is important for many societally beneficial projects. However, we also acknowledge that the history of American law enforcement has its origins in fundamentally racist endeavors such as slave patrols, and that these origins continue to influence modern law enforcement institutions. We also acknowledge that police forces in many of our own communities have actively demonstrated brutality and racism towards our communities, coupled with poor judgment in their acquisition and deployment of equipment. As such, we cannot in good faith trust these police forces with the types of robotic technologies we are responsible for researching and developing.

The strongest protection against future misuse is likely to be based in policy and legislation to provide necessary oversight and protection of civil liberties. We believe that the clearest path toward preventing this misuse is to (1) pass legislation requiring state and local governments to acquire informed public consent before acquiring robotic technologies, similar to Massachusetts Senate Bill S1358; (2) advocate for policies that decrease police departments' ability to acquire robotic technology, and that instead work towards community health, education, and affordable housing; and (3) where police acquisition of robotic technology cannot be curbed, advocate for appropriate training in the use of such technology, and for policing strategies that are oriented around community-oriented policing.

While we agree to advocate for such policy decisions, it is nevertheless important to recognize that policy decisions are ultimately out of our control, and that we must thus take whatever additional actions that are within our control. Accordingly, we additionally pledge the following actions.

1. We refuse to facilitate the execution or publication of research in robotics or Artificial Intelligence (AI) that is performed in collaboration with State and Local Law Enforcement agencies (or national agencies such as ICE and CBP).
2. We refuse to pursue or accept funding for research projects whose express goal is development of robotic technologies or AI technologies (e.g., nonconsensual face recognition technologies, predictive policing models, etc.) intended for use by such agencies.
3. For those of us who are affiliated with colleges and universities, we pledge to petition our universities to divest themselves of all ties with law enforcement agencies that have demonstrated brutality, prejudice, or white supremacist tendencies.