The British science-fiction writer Arthur C Clarke once made the point that advanced technologies would one day become ‘indistinguishable’ from magic – or, at least, would appear so to a visitor from an earlier age. Modern computing has, on one level, delivered great advances and innovative products that would fit Clarke’s ‘magical’ criteria. However, these innovations haven’t been delivered to all, with the majority of mankind not reaping their benefits.
At the moment, hi-tech products aren’t nearly as ubiquitous or ‘magical’ as they should be. With that in mind, the news that ‘Eugene Goostman’, a chatbox developed by Amazon software developer Vladimir Veselov, has become the first computer to pass the ‘Turing Test’, has caused quite a stir. The Turing Test measures a computer’s ability to exhibit behaviour that is indistinguishable from that of a human. The chatbox passed the test by convincing more than 30 per cent of a group of human participants that it was a 13-year-old Ukrainian boy. While many have dismissed this as a mere stunt, it reminds us, at least, of the potential of artificial intelligence (AI) for the future.
Some have voiced concern that this event proves AI is ‘going too far’, with Professor Kevin Warwick stating that this should serve as a ‘wake up call for cybercrime’. However, a little alarmism aside, this breakthrough does raise some longstanding philosophical questions about the difference between AI and the human mind. Are humans little more than complex machines with our CPU located in our skulls?
The short answer is: no. The conflation AI enthusiasts make between the mind and the brain ignores the fact that human consciousness, ‘the mind’, is only partly located in us as individuals. As Kenan Malik suggests in his book, Man, Beast and Zombie, the human mind is, in part, a social mechanism. The mind of an individual, Malik writes, is ‘best understood as a set of relationships between the individual and the social, cultural and historical networks he inhabits’.
AI should be continued to be developed; it may prove to have great practical applications and will only contribute to a better understanding of what it means to be human. But unless AI engineers begin to produce Eugene Goostmans that are capable of developing independent social and cultural networks, AI will never, in any profound sense, be able to mimic human consciousness. This is something worth bearing in mind, amid all the excitement at the Turing Test finally being beaten.