Our AI Overlords Will Save Earth, Says Prominent ScientistAlphaGo, the Go-playing computer program, is the start of telepathic superintelligences that will tackle climate change
James Lovelock, both distinguished scientist and founder of the Gaia concept of Earth, need only live to July 26 to be a centenarian. In his new book, co-authored with Bryan Appleyard, he sees the Go-playing computer program AlphaGo as the start of a new type of life that will save the planet, as he told New Scientist recently:
Specifically, about the new AI overlords: In his new book Novacene, James Lovelock says the creation of AlphaGo was the start of a new kingdom of life that will create and think for itself. He’s optimistic that this new kingdom of life will want to keep us around like we keep plants in gardens.“James Lovelock says artificial intelligence is the start of new life” at New Scientist
Penguin, his publisher for Novacene: The Coming Age of Hyperintelligence, explains,
James Lovelock, creator of the Gaia hypothesis and the greatest environmental thinker of our time, has produced an astounding new theory about future of life on Earth. He argues that the anthropocene – the age in which humans acquired planetary-scale technologies – is, after 300 years, coming to an end. A new age – the novacene – has already begun.
New beings will emerge from existing artificial intelligence systems. They will think 10,000 times faster than we do and they will regard us as we now regard plants – as desperately slow acting and thinking creatures. But this will not be the cruel, violent machine takeover of the planet imagined by sci-fi writers and film-makers. These hyper-intelligent beings will be as dependent on the health of the planet as we are. They will need the planetary cooling system of Gaia to defend them from the increasing heat of the sun as much as we do. And Gaia depends on organic life. We will be partners in this project.
Partners? At best. “ Pets” and “ plants” are also proposed as our roles. Lovelock does not believe that there are intelligent aliens out there, so he thinks that the needed intelligence must be manufactured by machine superintelligences on Earth.
As he explained to New Scientist,
“It was really kicked off by AlphaGo, the application of mathematical modelling in a much more constructive way than had been done previously. It’s not a logical cause and effect thing. The programme is in a sense choosing its own bits and pieces. If that is not the start of life I would like to know what is. And I see Mr Darwin hovering in the background there, thinking, ‘Right, yeah, now that’s going to evolve.’ […] The artificial intelligence today can think for itself. They will create themselves.”“James Lovelock says artificial intelligence is the start of new life” at New Scientist
Lovelock’s contributions to conventional science include an argon detector and an electron capture detector. When he first got involved with the idea of the Gaia hypothesis, it started out as science:
With microbiologist Lynn Margulis, Lovelock published a series of papers on the subject. In 1974, they developed a view of Earth’s atmosphere as “a component part of the biosphere rather than as a mere environment for life” (J. E. Lovelock and L. Margulis Tellus 26, 2–10; 1974).Tim Radford, “James Lovelock at 100: the Gaia saga continues” at Nature
But it seems to have become something more for Lovelock. By 1979, he was saying things like. “The entire range of living matter on Earth from whales to viruses and from oaks to algae could be regarded as constituting a single living entity capable of maintaining the Earth’s atmosphere to suit its overall needs and endowed with faculties and powers far beyond those of its constituent parts.”
His Gaia hypothesis swept through popular culture. What most people know today is a green goddess, an icon of panpsychism, as in Could Stars Be Conscious?, from a Gaia publication. Lovelock’s books went on to include Gaia (1979), Homage to Gaia (2000), The Revenge of Gaia (2006), and The Vanishing Face of Gaia: A Final Warning (2009).
Some colleagues have grumbled openly that it isn’t science. Lovelock does not worry. He awaits the Singularity. Our AI overlords, he thinks, will rescue Gaia because, unlike us, they will realize the truth:
Novacene picks up from that note of hope, and showcases another big idea. Gaia might, after all, be saved — by the singularity. This artificial-intelligence takeover, which so alarms many doomsayers, will be our redemption. Lovelock argues that increasingly self-engineering cyborgs with massive intellectual prowess and a telepathically shared consciousness will recognize that they, like organisms, are prey to climate change. They will understand that the planetary thermostat, the control system, is Gaia herself; and, in tandem with her, they will save the sum of remaining living tissue and themselves. The planet will enter the Novacene epoch: Lovelock’s coinage for the successor to the informally named Anthropocene.Tim Radford, “James Lovelock at 100: the Gaia saga continues” at Nature
We are, Lovelock tells Radford, but “parents and midwives” to the savior cyborgs.
Is he perhaps looking for something beyond science to save us all? If so, AlphaGo and its successors are likely to disappoint, says computer engineer Eric Holloway: “I wonder why he picks AlphaGo, versus all the other game playing AI out there. “I submit the humble matchbox as the new form of Novacene life.”
AI is well suited to winning games, which present precise lists of rules based on logical strategy, developed by humans. Modern AI is much faster and more complex than mechanical systems involving matchboxes but the fundamental limitation remains: Nature does not present a list of precise rules for success. Success requires constant adaptation to unknown and possibly entirely new circumstances.
That’s just as well. Lovelock’s new kingdom of life would not all be sweetness and light anyway, not if we go by his thoughts on Elon Musk’s proposed Mars mission, as told to New Scientist,
“I know a fair amount about Mars. I don’t think we’ll start colonies on Mars. I cannot think of a much more inhospitable place. I think Elon Musk is a very clever man, he must be, [otherwise] he wouldn’t be so rich. But to want to go and live on Mars is just about as crazy as you could be. He must hate people even more than I do.”
You said it would be better for Elon Musk to crash on impact
“Yes.”“James Lovelock says artificial intelligence is the start of new life” at New Scientist
See also: The flawed logic behind thinking computers: Part III (Eric Holloway)
Tales of an invented god
Featured photo: El-santuario-de-Gaea-Juan/mrgt, Adobe Stock