Recent startling successes in machine intelligence using a technique called ‘deep learning’ seem to blur the line between human and machine as never before. Are computers on the cusp of becoming so intelligent that they will render humans obsolete? Harry Collins argues we are getting ahead of ourselves, caught up in images of a fantastical future dreamt up in fictional portrayals. The greater present danger is that we lose sight of the very real limitations of artificial intelligence and readily enslave ourselves to stupid computers: the ‘Surrender’.
By dissecting the intricacies of language use and meaning, Collins shows how far we have to go before we cannot distinguish between the social understanding of humans and computers. When the stakes are so high, we need to set the bar higher: to rethink ‘intelligence’ and recognize its inherent social basis. Only if machine learning succeeds on this count can we congratulate ourselves on having produced artificial intelligence.
This is an interesting book to read. From the very start it is clear that it is an oppositional book, aimed at countering a certain point of view – namely that AI is moving as fast as some other advocates feel it is. This can make you instantly wary as you can sense the author has already made their mind up. The title too is of course making a clear statement too (though it took a few times with the book before we noticed this!) about the author’s feelings.
All of this said however the book then aims to dispassionately consider the claims made about AI and it’s potential to reach certain goals suggested by industry advocates and evangelists such as Geoffrey Hinton and Ray Kurzweil among others. The more we read this book, the more we enjoyed it’s arguments and attempts to explain the reasoning behind them. At the same time the author does acknowledge that no one anticipated the success of pure brute force in some of the recent AI achievements. Of course it doesn’t mean that the AI in question actually understands how to play chess or even Go!, but it can simply do it better, faster and smarter than any human can now. Regardless of this ‘how’, it is still clear that in this narrow form of AI, we have seen rapid massive unexpected achievements. This does illustrate that there are unknown unknowns which could still enable AI to achieve some goals faster than Collins believes may be possible. As Collin’s himself acknowledges he may just be plain wrong on some of his predictions simply due to the challenging nature of estimating how fast some areas of AI will progress. At the same time this is a good, useful book to read because, even as a non industry expert, Collins has still read deeply in this area, and consequently is posing some important, challenging questions. Having already experienced long periods of AI winters already this book provides a robust challenge to those techno solutionist optimists who see AI delivered solutions through overly rose tinted glasses.
This is a useful addition to the world of AI literature and a timely book to read as it raises important questions that do need to be considered.