1,000,000,000,000,000,000 calculations per second
Posted on 25th March 2019
The importance of tech is evident in the increasing amount of space the mainstream press gives it and the burgeoning number of journalists covering a multitude of related subjects. Of course, the accompanying politics of tech – the fines for Google, the privacy issues around social media use, etc. – all help sell newspapers, or rather help slow down the decline of their print editions, but there is no doubt that every week there are lots of entertaining stories and often illuminating comment and opinion. This, in my opinion, is all to the good. Technological change, from the wheel to the steam engine to the world-wide-web, has always driven us, so it’s great to learn what’s happening in the broader world. Here are four stories that particularly caught my attention last week…
Firstly, for all those who remember the Hitchhiker’s Guide to the Galaxy and the Deep Thought computer, it’s instructive to learn just how powerful computers are nowadays. The US has announced that it’s to create the world’s most powerful supercomputer capable of 1 quintillion calculations per second. And in case you were wondering, that’s 1,000,000,000,000,000,000 calculations per second. Named ‘Aurora,’ it will be approximately seven times faster than today's most powerful supercomputers. Crucially, if it succeeds, it could allow the US to take the lead in its high-stakes computing race with China. I also think it would probably have a decent chance on Mastermind. Whether the answer to the ultimate question is 42 remains to be seen though.
Secondly, and bearing in mind Aurora’s computing power, I rather liked the poll that showed c. 33% of British voters would prefer AI to lead the country than a politician. Now before we get into jokes about how we’re already led by robots, I should point out that this was a Europe-wide poll and, perhaps surprisingly given our current travails in the UK, it’s Holland where most voters (43%) would like the computers to run the country rather than the politicians. It’s also worth noting that the same poll showed that 70% of the respondents think that, on balance, technology will cause more harm than good in the next decade, while two-thirds are concerned that our social lives will be dominated by virtual, rather than human, interactions. People were also worried about robots taking jobs, but not, obviously, that worried about them taking politicians jobs.
Following on from that, the BBC reported on the ‘murder’ of a robot in the US. This happened back in 2015 on the streets of Philadelphia and was a little robot called Hitchbot. As the BBC report said, “The ‘death’ raised an interesting question about human-robot relationship - not so much whether we can trust robots but whether the robots can trust us. The answer, it seems, was no….Hitchbot has now been rebuilt, at Ryerson University, in Toronto, where it was ‘conceived’.”
The final story I picked up on was also from the BBC’s tech pages, where it is reported that (as we’ve noted before on the Be-IT blog), there are concerns over potential bias in algorithms. In particular, because they are programmed by fallible humans, there is a danger they’ll have gender and race biases. Consequently, the Department for Digital, Culture, Media and Sport has asked an independent watchdog, the Centre for Data Ethics and Innovation, to investigate the algorithms used in the justice and financial systems. However, some services using AI already, such as predictive policing, will continue to do so. More pertinently, for recruiters such as Be-IT, the CDEI will also look at potential bias in algorithms used in recruitment to screen CVs and influence the shortlisting of candidates.
Matt Druce, Be-IT
Posted in News, Recruitment News
.. Back to Blog