This seemed fair enough at the time, if admittedly challenging. What I hadn’t appreciated, though, was the controversial nature of such a claim. For sure, in that piece I referred to Yoshua Bengio’s argument that we don’t understand human experts and yet we trust them, so why should we not extend the same degree of trust to an expert computer (Pearson 2016)? But it transpires this is quite a common argument posited against claims that systems should be capable of explaining themselves, not least among high-level Google scientists. For example, Geoff Hinton recently suggested in an interview that to require that you can explain how your AI systems works (as, for example, the GDPR regulations do) would be a disaster:
In his book Homo Deus, Yuval Noah Harari rather randomly chooses archaeology as an example of a job area that is ‘safe’ from Artificial Intelligence:
The likelihood that computer algorithms will displace archaeologists by 2033 is only 0.7 per cent, because their job requires highly sophisticated types of pattern recognition, and doesn’t produce huge profits. Hence it is improbable that corporations or government will make the necessary investment to automate archaeology within the next twenty years (Harari 2015, 380; citing Frey and Osborne 2013).
It’s an intriguing proposition, but is he right? Certainly, archaeology is far from a profit-generating machine, but he rather assumes that it’s down to governments or corporations to invest in archaeological automation: a very limited perspective on the origins of much archaeological innovation. However, the idea that archaeology is resistant to artificial intelligence is something that is worth unpicking.
In 2014 the European Union determined that a person’s ‘right to be forgotten’ by Google’s search was a basic human right, but it remains the subject of dispute. If requested, Google currently removes links to an individual’s specific search result on any Google domain that is accessed from within Europe and on any European Google domain from wherever it is accessed. Google is currently appealing against a proposed extension to this which would require the right to be forgotten to be extended to searches across all Google domains regardless of location, so that something which might be perfectly legal in one country would be removed from sight because of the laws of another. Not surprisingly, Google sees this as a fundamental challenge to accessibility of information.
As if the ‘right to be forgotten’ was not problematic enough, the EU has recently published its General Data Protection Regulation 2016/679 to be introduced from 2018 which places limits on the use of automated processing for decisions taken concerning individuals and requires explanations to be provided where an adverse effect on an individual can be demonstrated (Goodman and Flaxman 2016). This seems like a good idea on the face of it – shouldn’t a self-driving car be able to explain the circumstances behind a collision? Why wouldn’t we want a computer system to explain its reasoning, whether it concerns access to credit or the acquisition of an insurance policy or the classification of an archaeological object?