Being a niche question about Gen AI I felt inclined to ask Chat GPT its views on the question. Purely for the purpose of this essay, I have used some of the ideas given by Chat GPT but not all. AI can be used in economic policy making to conduct data analysis, predict certain economic trends and even make decisions on what policies need to be implemented. AI is useful for all aspects of life ,not just economic policy making, but there can be some negative ethical implications such as the possible replacement of jobs, reduction in public participation/democracy and the lack of the ability to address bias. However, the scale of the impact can differ and may not be as serious as first assumed.

It is widely accepted that AI will replace jobs in the future and is already doing so. The automation of jobs in the economic policy making area include data analysts, economic forecasters and even researchers as surveys and data collections can now be mostly done online. Unemployment may increase leaving many highly skilled economic policymakers jobless. This could lead to further income inequality as those who are equally as skilled but in areas that are less likely to be taken over by Gen AI: certain doctors, coders, scientific researchers e.t.c , may remain on a high salary whilst there are more people unemployed and on a low income. However, it is extremely important to consider whether the policymakers will have their jobs fully replaced or only half replaced. They may have less work to do as Gen AI is more efficient and can do administrative jobs that are part of a larger role an individual has. This could mean with their working hours reduced they are paid less or needed less often. The other part of the job such as the human judgement and ethical considerations of society an economic advisor needs to consider, cannot be done by Gen AI. For example, consider a decision on whether to remove welfare benefits for unemployed people in order to free up government funds for the NHS ,was needed to be made. The consideration of the people who may now struggle and suffer from poverty with worsening mental and physical health, may not be taken into account by AI. Yes, AI will take into consideration all the economic implications but it may struggle to consider the psychological impact certain decisions will have and be potentially unable to assess the trade offs between welfare and what is best for the economy. As a result, AI is not able to replace all economic policy making  jobs completely, but will have some sort of negative effect. 

When making economic decisions these can be proposed wat multiple different times such as during an election cycle by the winning party, by think tanks or even in response to sudden shocks such as war or a recession. Many policies are made by the government solely without the public having a say, with the exclusion of referendums. Yet decisions are usually explained fully by the prime minister or members of parliament so the public understand how this will affect them. Yet with the use of AI, the so called ‘Black Box problem’ may occur. This is where the public may find it difficult to understand how the AI has come to certain conclusions on what decisions to make and creates a lack of transparency .This may cause the public to lose trust in the government, especially if it is not declared that AI has been used to help policymakers reach a decision/create a policy. However the issue also arises that with generative AI it is difficult to tell if the answer is accurate thus it is susceptible to making mistakes and providing false information. The question of who is accountable for these mistakes then arises but it does of course depend on the context and the purpose of the AI. For example, a policy maker may use incorrect information that they received from AI on conclusions from a survey on their constituencies grievances. This could lead to ineffective or factually incorrect policies from being made which may affect the economy in some way. However, it has been debated whether the policymaker is held accountable or the AI. Some may argue that with AI itself, it cannot be held accountable as it is not a person, so the programmer should be. Yet, generative AI does not have a set programmer and is a mixture of thousands of different databases so it is difficult to hold it accountable.

Finally a slightly disappointing point is that there is bias in AI, which reflects human bias. This is an issue that is difficult to combat as it involves technical data science, but without it being addressed it stops people’s ability to take part in society and the economy. Economic policies that are made based on one race or gender are unequal and any other discriminatory feature that AI may present a bias to, goes against the equality act in the UK. Despite there not being any regulation associated with AI, bias should be minimal as it will ensure a more fair and accurate outcome form the use of AI. Attending a Global Tech Advocates Artificial Intelligence event, I listened to Professor Tim Dingle explain his business where he uses AI to create personalised training for workplaces and school, by identifying their learning styles. Bias is overcome here by creating multiple choice questions when asking how one learns to avoid bias of certain trigger words associating with a certain learning style. Clearly there are ways of avoiding this but when it comes to economic policymaking, often bias appears when data is presented especially when there was racial minorities involved. Often AI can rely on historical data which leads to less focus on these minorities and more on Caucasians.

Overall, despite its ethical implications, Generative AI will become more prominent in the future with it being used in all different sectors as well as in economic policymaking. Overtime, as AI is used more often there will be the chance that these ethical issues can be solved yet this is unpredictable and humanity may have to live with this as AI takes over out world.