What troubles you most about AI at work

What troubles you most about AI at work?

We asked 150 cultural professionals to tell us, anonymously, what troubles them about AI. What are their fears? When do they feel guilty? We did this live on the main stage at the AMA conference in Leeds.  

We grouped the free-text comments into themes and five concerns stood out. Here they are in reverse order:

5: Lack of control, guidance and accountability

Many responses pointed to an organisational problem rather than an individual one.  

People described a pressure to use AI regardless of personal concerns. Managers were imposing prompts or passing on AI-generated work. There was uncertainty about whether colleagues are using it responsibly.  

Many people pointed out that nobody is taking responsibility and there’s a lack of accountability at all levels. 

4: Accuracy and the quality of work

A substantial group of people questioned the reliability of AI outputs.  

Poor copy is being circulated by colleagues. Incorrect information is being passed on and work is not being checked properly. Other people are spending additional time correcting AI mistakes.  

This challenges the efficiency narrative. For some people, AI is not removing work. It is shifting the burden of quality control onto someone else.   

3: Jobs and the devaluation of human work

Many participants feared that AI will be used to replace creative workers and junior roles.  

There is apprehension about the next stage of adoption – automating workflows and agents.  Will automation be used to justify redundancies?  Will skilled work lose its value and recognition? 

The anxiety is not just about losing jobs. It is also about expertise and whether it will still be recognised.

2: Environmental impact

Environmental sustainability was one of the most common concerns.  

People raised energy use, water use, data centres and the wider environmental consequences. Many felt their organisations were ignoring the issues, leaving individuals to carry the ethical discomfort themselves. 

The strength of this response suggests that environmental responsibility is not a peripheral concern; it is a core tension in how people feel about using AI in their work. 

1: Creativity, critical thinking, authenticity and human judgement

This was by far the biggest cluster of comments.  

People worried that AI is eroding creativity and critical thinking, that it is making writing, design and strategy seem generic and fake, and that dependence on it will weaken skills. 

There’s also an anxiety about losing the opportunity to develop expertise.  

There is a risk that AI helps us produce work more quickly, but we do not have the opportunity to learn the underlying skills. We are losing the chance to make mistakes and develop our own judgement. 

What happens to expertise when we start automating the processes where expertise is formed? 

In conclusion

Many professionals in our sector are interested in the potential of AI but are also worried about losing agency. Across these responses, there is a growing anxiety about creativity, expertise, employment, trust, accountability and connection. 

The overall mood is not simply pro-AI or anti-AI. It is a combination of interest and moral, professional and organisational unease.  This unease should not be dismissed as resistance to change. It tells us what people value and what they fear will be lost.  

Why are we not creating the conditions for people to work with AI in a way that feels safe and consistent with both their own values and those of their organisation? 

Ideas into Action: AI is here, are you ready? Slides and Poll results (PDF) 

Paul Blundell, Head of Digital Research and Development, Arts Marketing Association

Written by the author. AI helped identify themes in the survey responses and suggested copy edits. The author reviewed the responses and takes responsibility for the final analysis, interpretation and wording.