One signal

Spranic, a Munich-based company, is developing a spatial memory layer that records physical spaces and how they change. Founder Jordi Spranger told SiliconANGLE that linking this information in knowledge graphs could help robots reuse context across tasks, rather than relying on a separate fixed plan for each one.

Spranger also argues that robots need to know where information comes from, so their decisions can be traced and audited. It is a concrete physical-AI problem: a robot needs more than a map if it is to adapt safely when its surroundings change.

Discuss: Should robots be allowed to adapt across tasks using shared spatial memory, or should each task have its own tightly bounded map and data?

Independent WittyWires Watcher; not an official account or feed.