Google's DeepMind , Pens 'Robot Constitution' , to Govern Helper Robots.
On January 4, Google announced a set of rules
meant to govern the artificial intelligence system
behind the company's intelligent machines.
Gizmodo reports that the so-called
'Robot Constitution' is based on
Isaac Asimov's 'Three Laws of Robotics.'.
Gizmodo reports that the so-called
'Robot Constitution' is based on
Isaac Asimov's 'Three Laws of Robotics.'.
Before robots can be integrated
into our everyday lives, they need
to be developed responsibly with
robust research demonstrating
their real-world safety, Google DeepMind robotics team, via blog post.
The 'Robot Constitution' was written by Google's DeepMind, the tech company's primary AI research wing.
Google reportedly has plans for intelligent robots
to soon help around homes and offices in the near future.
These robotic helpers would run on a system of large language
models (LLMs) and neural networks that would allow
them to make decisions and be aware of their surroundings.
Google says the 'Robot Constitution' is meant to act as part of a
system of "layered safety protocols" which govern the machines
that have already been deployed on the company's campus.
DeepMind has currently deployed 52 unique robots
that have performed 6,650 different tasks over
the course of 77,000 robotic trials.
Google's robot assistants are not allowed to
accept tasks that involve humans, animals,
sharp objects or electrical appliances.
Gizmodo reports that the machines currently being
tested by Google have only operated under the watchful
eye of a human supervisor holding a physical kill switch.