I got to thinking again about “The Singularity” when I read this week about robots teaching other robots as part of some research at MIT’s Computer Science and Artificial Intelligence Laboratory.
The technological singularity, often just simply called the singularity, is the idea that the invention of artificial superintelligence (ASI, a step beyond AI) will result in very rapid technological growth and then dramatic changes to human civilization.
The term was popularized by sci-fi writer Vernor Vinge in his 1993 essay “The Coming Technological Singularity” but goes back earlier to the mathematician John von Neumann, who spoke of “ever accelerating progress of technology and changes in the mode of human life, which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.”
Futurist Ray Kurzweil wrote The Singularity Is Near: When Humans Transcend Biology in 2005 as a sequel and extension to his previous books, The Age of Intelligent Machines (1990) and The Age of Spiritual Machines (1999). Kurzweil predicts an exponential increase in technologies like computers, genetics, nanotechnology, robotics and artificial intelligence, and then once the Singularity has been reached, machine intelligence will be infinitely more powerful than all human intelligence combined.
Kurzweil and others see the next phase to be when intelligence moves beyond Earth until it saturates the universe. Some people say that the true Singularity is the point at which machines intelligence and humans merge.
In the past decade, some famous folks like Stephen Hawking and Elon Musk expressed concern that full artificial intelligence could result in human extinction.
The Singularity’s potential benefit or harm to the human race is being debated.
But my current interest is this idea of the machines using “social learning.” In your lifetime, you have probably learned more by observing or interacting with others than you learned in any formal “school” setting. This kind of socially acquired knowledge is different from what we learn on our own, or in a classroom.
I guess I shouldn’t be surprised that robots would some day also learn this way. The robots at MIT are learning from one another. A robot is programmed with a knowledge base so it has information about how to interact with objects such as door handles. This knowledge base helps the robot navigate the constraints of the world, such as the physical necessity of having to turn a handle before pulling a door open. A human “teacher” only needs to demonstrate the action once. Then, the robot can pass its knowledge on to other robots.
Maybe that doesn’t fit you definition of teaching. True. This transfer of skills between robots still needs intervention from a human. For now…
What they are working on at MIT is demonstrating tasks to one robot that can then transfer its skills to other robots that are different. Others with different body shapes and strengths can use the skills in other ways. Their goal is independent social learning in robots – cultured robots.
If we reach the singularity – or is it that the machines reach it? – they will no longer need us. Does that make you feel hopeful or hopeless about the future?