AI challenges mean human experts have never been more important – The Irish Times
“We’re fed up with experts!” This famous phrase, uttered by Michael Gove during the 2016 Brexit debate in the UK, summed up the mood of many who were tired of being lectured by people distant from their daily lives rails.
This sentiment was also reflected during a Brexit Town Hall debate in Sheffield, when an economist spoke about the potentially negative impact that leaving the European Union would have on UK GDP, where a voice from the crowd challenged: “The is all very well when you talk about your GDP, but what about our GDP?”
While there have been many excellent contributions from scientific and public health experts during the pandemic, there has been evidence of growing public distrust and even rejection of scientific evidence and research. This is reflected in the discourse on climate change and biodiversity loss.
[ Artificial intelligence: the time for funny pictures is over ]
As we increasingly see evidence-based research being rejected, this has come with a growing distrust of those believed to be in control of that expertise.
Populists across Europe are increasingly basing their campaigns on what they market as “common sense”, in sharp contrast to boring, evidence-based reports.
Historically, when governments made a decision, after appropriate consultation or fact-checking, they were relayed through established channels such as RTÉ, The Irish Times or other traditional media. Similarly, when a scientist makes a discovery, it is peer-reviewed and published in relevant journals.
Today everyone can communicate their views and “facts” through digital channels. They don’t need to be fact-checked or peer-reviewed. The mainstream media or “so-called academic experts” have no control over what might best be described as “alternative truths.” In order to gain access to reliable fact-based journalism or scientific research, much of it is hidden behind a paywall.
And if it is not enough that those who risk their lives to make facts and truth count are questioned by those who see no value in evidence-based research, artificial intelligence and machine learning will replace our expertise as humans.
In his book Homo Deus (2016), Yuval Noah Harari asks a question at the end: “What happens to society, politics and daily life when unconscious but highly intelligent algorithms know us better than we know ourselves?”
[ The Irish Times view on the future of AI: time to pause ]
A great deal of global power has shifted to those who own and develop the complex algorithms that govern so many parts of our lives today.
While in the past those in power could direct the development of technology and its use based on expert advice, this role is increasingly in the hands of a small group of mostly men who tend to share broadly similar worldviews.
In the 18+ months that we at the Oireachtas Media Committee have been pre-legislating on what later became the Online Safety and Media Regulation Act, it has become increasingly difficult to attempt to regulate a digital environment that is becoming so quickly changed clear.
The convergence of new technologies should force more public and political debate on issues such as an ethical framework for the use of artificial intelligence; How far should we go with genetic modification of humans, animals and food? do we need some kind of geneva convention for cyberspace? and how might a non-aligned country like Ireland respond to a state-sponsored cyberattack?
The advent of machine learning capable of writing thoughtful articles poses an enormous challenge for universities, not only in terms of how students are evaluated, but also in terms of whether we need to invest in expertise from academic research if machines can do this for us. No longer “we have enough of experts” but “we don’t need any more experts”.
The convergence of technologies poses challenges, opportunities and questions for science, politics and all of us.
In medicine, would you trust a 100 percent accurate AI-enabled machine to perform your surgery or the human advisor, even though the latter would be more likely to make a mistake? Who should decide?
AI-enabled drones are now able to target and kill people. There is debate in defense circles about at what stage human intervention should occur: once the drone has identified a suspect, should it make the decision to execute, or should it remain in human hands?
I may have painted a dystopian society in which the human expert is challenged both by those unwilling to engage in evidence-based research and by technologies that may go further than we ever imagined.
But the role of human experts in society has never been more important. They must stand up for truth and for learning and confront those who seek to undermine research and academic endeavors. They must also lead the debate on how we as a society adapt to a world of quantum computing, machine learning and big data.
[ A rogue artificial intelligence is no longer science fiction ]
Human experts must continue to value ethics and social responsibility. But they are not allowed to conduct this discussion only in a university context. You must contribute to the public debate.
The challenge for everyone in the research community is to explain in an understandable way why and how their research can make a difference and why it is important for research to influence policy.
It also means understanding the concerns and fears of those who think “experts” are after them and trying to convince them of the importance of research.
As part of this contract, the government must value this research and expertise and ensure that our education system supports both professionals and learners in the challenges and opportunities that the digital century will bring.
Every research strategy must always put a premium on research excellence. We can never have enough experts.
Malcolm Byrne is Fianna Fáil’s spokesperson for further and higher education, research, innovation and science. This is an abridged version of a speech he recently gave to the Royal Irish Academy