The Netherlands is looking forward to a coalition agreement and a new cabinet and in the meantime is preparing for the municipal elections on 16 March 2022. But the question is whether digitization and the use of smart systems (artificial intelligence) are a topic of discussion. And that is striking. After all, countless municipalities are rolling out ambitious plans for smart cities. Cameras in city squares and local industrial estates are getting smarter; some are already equipped with facial recognition.
Digitization has an enormous influence on society, but is not high on the political agenda. Even the Allowances affair has not resulted in people in The Hague seeming to be lying awake with systems that are in charge, instead of people. And even when it comes to the opportunities and risks of digitization for the earning capacity of our country, politics remains remarkably quiet.
There are positive signs. For example, the House of Representatives now works with a Standing Committee for Digital Affairs. And informateur Hamer spoke during her explorations with various representatives of the digital sector. Still, more is needed. More than with previous digital technologies, all kinds of values are now at stake. It is time for government and government to get serious about managing the upcoming fundamental changes.
Last week, the Scientific Council for Government Policy (WRR), where we are chairman and project coordinator AI, therefore presented a report about the influence of artificial intelligence (AI) on society. Central to our advice is the idea that AI is a system technology, comparable to electricity, the combustion engine and the computer. This means that such a type of technology will have a long-term and large-scale impact on the economy and society, with unforeseeable effects at the moment.
How can the government ensure that AI is properly embedded in society? To this end, we distinguish five social tasks.
First of all, the government must do more to create a realistic image of AI. This is often distorted by misplaced fears and overwrought expectations. AI is not an evil robot – nor is it the answer to everything. A realistic image helps to avoid disappointment and can ensure that citizens dare to embrace the good sides of the technology. Algorithm registers, for example, can provide the general public with insight into AI use, thereby stimulating the debate about AI and working towards realistic expectations.
Second, good use of AI means looking beyond the technology itself. For example, to allow the car to drive safely, adjustments in the environment were necessary at the time. Think of (motorways) roads, signage, inspections. Similarly, AI requires changes in the workplace, in employee skills, and the broader technical environment in order to ‘read’ it.
Such adjustments must therefore be made equally. After all, otherwise the technology cannot and will not work properly in practice. The following applies: not everything can be done at the same time. So priorities have to be set. It is therefore important that the Netherlands now focuses on its own ‘AI identity’, for example in areas where we have a global economic position to defend, such as logistics and agriculture. But also in areas where there are major social interests, such as healthcare, education and sustainability.
Also read this editorial: AI requires an overarching vision for the future
Social counterplay
The third and fourth task that we distinguish revolve around the need for better social counterplay and a more strategic view of the necessary regulation. More counterplay is needed to prevent the interests of large companies from being dominant in the development of AI and to exclude certain population groups. A more strategic view of regulation should avoid AI legislation being reactive, ad hoc and issue specific. A legislative agenda for the digital environment is also needed to prevent the dangers of mass surveillance and concentration of power at a limited number of tech companies from overshadowing the positive contribution of AI.
Finally, the international position of the Netherlands is at stake. As a fifth task, the WRR therefore advises the government to guarantee the earning capacity and safety of our country, even in a world in which AI is becoming increasingly important. AI innovation is a tool in the hands of dangerous players.
Attention must also be paid to the important choices that are currently being made in the field of standardisation. When you consider previous system technologies, you realize how crucial standards are. And with it the importance of being able to influence the choice for a certain standard. In collaboration with other European countries, the Netherlands must develop AI applications and a good infrastructure for AI and commit to global rules and standards based on what we call integrated ‘AI diplomacy’. With the five tasks, we provide a framework for properly embedding AI in society. The message of our report received a lot of attention, but that is not enough. A political translation of the advice is urgently needed. Otherwise we will be left with problematic imagery, poor application, exclusion of population groups, a government that is losing its grip on developments and has a weaker position on the international stage.
A version of this article also appeared in NRC Handelsblad on 20 November 2021
A version of this article also appeared in NRC in the morning of November 20, 2021
#Dont #artificial #intelligence #run