Some optimists, like technologist Nick Bostrom, believe AI is: “The last invention the human race will ever need”
However, given these are early AI days, fundamental questions still being asked include:
- How can we make sure robots’ only purpose would be to serve humans and our environment and make human lives better?
- Should robots make decisions on their own without approval from humans?
- Should we allow robots to fight autonomously in future wars?
- What will most humans do if most of their work is done by robots:
- Be idle or creative?
- Write, paint or compose better?
- Create new things, travel to new places and/ or make the environment cleaner?
- Enjoy more time in leisure and recreation?
- In particular, what will the many uncreative, lounge lizards do given having so much more free time will be a big challenge not only for them but society as a whole?
In addition, some high-profile players are most pessimistic about the future role of AI:
- Elon Musk, the US billionaire, developer of PayPal, owner of Tesla electric cars and founder of the Space X interplanetary rocket company says:
- “We need to colonise Mars so we’ll have a bolt-hole if AI goes rogue and turns on humanity”
- “AI scientists may have perfectly good intentions but still produce something evil by accident, capable of destroying mankind”
- “AI will be more dangerous than North Korea”
- Professor Stephen Hawking said:
- “The development of full AI, which can match or surpass humans, could spell the end of the human race – it would take off on its own, and re-design itself at an ever-increasing rate
- “AI has the potential to be the worst event in the history of our civilisation”
However, Microsoft co-founder Bill Gates pooh-poohs their pessimism and says:
- “AI will mean longer vacations”
- “Machine learning will make humans more productive and therefore able to accomplish the same amount of work in less time – that’s a good thing”
- “The purpose of humanity is not just to sit behind a counter and sell things – more free time is not a terrible thing”
But then Niall Ferguson of the Hoover Institution, Stanford, pours cold water on Gates’ optimism, claiming in a Sunday Times article: “The more machines learn, the less we shall grasp” – as machine learning replaces human judgement, we shall find ourselves as baffled by events as our pre-modern forefathers were – we shall no more understand the workings of the machines than they understood the vagaries of nature, citing modern flash financial crashes versus past flash floods
Worryingly, former Google chairman Eric Schmidt says: “Even the best Silicon Valley software engineers no longer fully understand how their own algorithms work”
The last words here go to Ferguson:
- Machines can already ‘teach themselves’ and this deep learning goes deeper than our paltry human minds can fathom
- Mankind thus stands on the threshold of a new era – a new mental era – where the sum of human happiness will be increased by deep learning
- The problem is the sum of human understanding may also end up being reduced