AI risks: call for expert assessments
Two expert assessments are part of the sub-project “Systemic risks of artificial intelligence”:
Systemic environmental risks of AI
The first tender is related to the environmental impacts of artificial intelligence technology. The development and roll-out of AI has accelerated dramatically in recent years and shows no signs of slowing down. At the same time, AI consumes vast amounts of resources and may have complex secondary effects that pose systemic risks to the environment. In this tender, the project is therefore looking for experts that can assess the causes, pathways, extent, and governance measures for those risks. To the tender (PDF)
Systemic risks of interacting AI
With the second tender, the project is looking for experts on the state of the art of interacting AI. Recently, agentic AI has attracted the attention of major AI companies. AI agents are expected to independently assist in or autonomously perform specific tasks they were previously trained on. Without disregarding the lessons learned from previous waves of automation, experts here are also called to assess the causes, pathways, extent and governance measures for systemic risks of this latest iteration of interacting AI. To the tender (PDF)
A further expert assessment has been tendered in the sub-project “Uncontrollable artificial intelligence: An existential risk?”
Existential risks of AI: technical conditions
Discussions about possible existential risks associated with AI are increasingly finding their way into scientific and political debates. The focus is on speculative scenarios that describe AI as potentially uncontrollable, potentially constantly self-improving, potentially misanthropic and a threat to the survival of humanity. The aim of the expert assessment is to examine the conditions that need to be met for such scenarios to occur, and the extent to which these conditions have already been met. To the tender (PDF)
The deadline for submitting proposals on all expert assessments is 31 March 2025.(06.03.2025)
Further information:
- Subproject page Systemic risks of artificial intelligence
- Subproject Uncontrollable artificial intelligence: An existential risk?