Eric Schmidt: AI should have verification systems to avoid 'undesirable outcomes'
Eric Schmidt: AI should have verification systems to
avoid 'undesirable outcomes'
By Tom Warren on
December 22, 2015 05:02 am
Eric Schmidt has long been a prominent supporter of
research and investment into artificial intelligence. The Google (Alphabet)
chairman has been involved in the company's self-driving car and predictive
search engines, and previously warned we shouldn't fear a future of AI. In a
new op-ed in Time Magazine, Schmidt praises the promise of AI, but warns it
"will require the right approach."
Schmidt calls for the makers of AI to follow three
guiding principles. The first is that "AI should benefit the many, not the
few," and always ensure that any creation "aims for the common
good." Schmidt also wants AI development to be "open, responsible,
and socially engaged." Challenges and questions will undoubtedly arise as
AI becomes a lot more powerful, and Schmidt wants an open and collaborative
conversation.
SCHMIDT WANTS TO AVOID UNDESIRABLE OUTCOMES
The third principle involves avoiding "undesirable
outcomes." There are many fears around AI replacing jobs and generally
taking over the world. Many of those concerns may be misguided, but Schmidt is
calling for verification systems to ensure an AI system "is doing what it
was built to do." The thought of an AI system going rogue or out of
control will add fuel to many existing fears, but Schmidt believes AI is just technology
designed to improve our lives and controlled by humans:
"We are building tools that humans control. AI will
reflect the values of those who build it. Ultimately, our dream for AI is to
give people more choices about how they live their lives. Under our control, it
can take the drudgery out of work and free up many more hours for creative
pursuits. And applied collaboratively, AI could help bring about solutions to
the world's most complex problems."
Comments
Post a Comment