The BBC reported Google’s DeepMind has been defeating world-champion human competitors in Go and Chess – all without knowing the rules of play prior to the engagement.
In 2016, we introduced AlphaGo, the first artificial intelligence (AI) program to defeat humans at the ancient game of Go. Two years later, its successor – AlphaZero – learned from scratch to master Go, chess and shogi. Now, in a paper in the journal Nature, we describe MuZero, a significant step forward in the pursuit of general-purpose algorithms. MuZero masters Go, chess, shogi and Atari without needing to be told the rules, thanks to its ability to plan winning strategies in unknown environments.
DeepMind website
This is an impressively scary development in computing intelligence from IBM’s Deep Blue. Learning game theory and defeating humans at strategy games is just one aspect of AI, another is understanding natural language as developed by OpenAI. What does the Turing Test mean in an age when millions of people are falling in love with an AI chat-bot?
Science fiction has admonished us to beware the dangers of artificial intelligence – the creation of a machine intelligence that will surpass our own and often concludes that we are unnecessary.
Adrienne Mayor, however, provides compelling evidence that technological advances have been historically and universally used most effectively to solidify totalitarian rule (see the related observation in Privacy).
The thesis of her Gods and Robots (2018) is summarized in this lecture for the Long Now
The advance of technology appears inevitable and (after a brief Luddite opposition) is widely embraced as a good thing. It makes life easier. But easier isn’t the same as better.
The human species needs stressors and challenges – without gravity muscles atrophy quickly, loading the skeletal system keeps bone density higher, the brain craves information, stimulus, and problems to solve. Biologically we need to strive against and toward.
Protective Arts 