Artificial intelligence tends to be described as though it were happening in one place at a time. It is not. It is happening in a great many places at once, in different languages, at different speeds and for quite different reasons. Any account of the British work in the field reads better when it is set beside the rest of the world rather than held up on its own, and Nicole Junkermann finds the comparison clarifying rather than deflating.
What Britain brings
The British contribution has always leaned towards the theoretical end. There is a long local habit of asking what a machine can be said to know, and of insisting that the claim be testable rather than merely impressive. That habit is considerably older than any of the equipment now used to pursue it, and it is the part of the work that travels furthest, because a good question keeps its value long after a particular system has been retired.
The second thing the country brings is spread. Research here is not gathered into a single valley. It is distributed across university towns from the south coast to the far north, which means the work happens in ordinary places, alongside ordinary local economies, and the people doing it tend to stay where they are rather than all migrating to one address.
What comes the other way
None of this runs in one direction. A great deal of the engineering that makes modern systems practical has come from elsewhere, and increasingly from countries that were nobody's idea of a centre twenty years ago. India is the clearest instance: a very large number of capable engineers, a strong appetite for tools that solve plain everyday problems, and a set of languages that obliges anyone working there to think past English from the first day. That last discipline is one Britain benefits from importing.
The comparison worth making
The temptation is to score countries against one another, which is both unpleasant and uninformative. The more useful question is what each place has turned out to be unusually good at, and whether the results are reaching people who can actually use them. On that measure the picture is encouraging, because the distance between a research result and something a small organisation can run for itself has been shrinking steadily for several years.
Where it leaves us
Nicole Junkermann is an observer here rather than a participant, and the observation is simply this. The interesting version of British artificial intelligence is the outward looking one. A country of this size has never had the option of working alone and has rarely tried to. Its strongest periods have all involved absorbing ideas from wherever they happened to come from and then adding something of its own.
Which is a reassuringly unglamorous conclusion. No single place holds the whole of this subject, and the countries that get the most out of it are usually the ones that stay curious about everybody else.
Visit the Nicole Junkermann homepage, read more about Nicole Junkermann, see the full biography of Nicole Junkermann, or return to the journal.
