Artificial Intelligence (AI) is the next defining evolution in human technology, and we are already seeing hundreds of thousands of companies adopting it across the world. The simple fact is that if you don’t take advantage of AI, you will be left behind. The media industry is one of the most prevalent examples of this. Over the past five years, we have seen AI used for content creation, editing, subtitling, advertising, composition, analysis, and much more. The number of AI applications within the media industry is only going to skyrocket over the coming years, so how do you seamlessly implement AI into your workflow? We sat down with Dr. Alexandra Dobra-Kiel to find out.

Fortunately, there is a clear blueprint for implementing AI into the workflow, with an emphasis on finding a solution that helps your specific company. The mistake most people make is confusing AI implementation with AI adoption when really they are completely different things. Implementation focuses on the technical part while adoption focuses on how you take your current employees on board with you. AI can be quite an anxiogenic topic and most employees’ first question is ‘will I lose my job?’ So if companies only implement AI without also adopting it, they create a culture of anxiety among their workforce, which only builds over time.

Any sector that involves large volumes of data, repetitive tasks, and a lot of prediction is very likely to be able to benefit from AI. Fast forward five or ten years and this will be every single company within every single sector. The industries that would immediately benefit from AI implementation right now are manufacturing and finance. That’s not to say that the media industry and a great number of other industries cannot benefit immediately.

The first stage in your adoption journey of AI  should revolve around motivation. Educate your employees about AI and how you intend to use it. Explain how it will enrich the company and open opportunities for employees. Motivate them and dispel any worries surrounding job security. 

The second stage in your adoption journey of AI should revolve around proficiency. What are you actually going to do with AI? How will it improve the way you do business? There are so many applications for AI within the media industry, but it is important to come up with a detailed and thorough plan regarding how and why you are going to use it. Walking into the world of AI blind is a surefire way to make mistakes and fall short of using it as effectively as possible.

The third step in your journey to AI should revolve around the topic of ethics. Ethics is not recommended as the final step because it is the least important, but because it serves a better purpose after you have completed your motivation and proficiency stages. Starting with ethics can be rather restrictive. Only once you fully understand something can you actually understand the complete range of complex ambiguities and dilemmas surrounding it. Having ethics as your final step also serves as a sufficient failsafe to any social shortcomings of your plan. You have figured out whether you can do something, but then comes the idea of whether you should do it. Those are very different things.

“We need companies to focus on ‘what we can be’ rather than ‘what we can’t be’. It’s about enabling the good uses of AI rather than just restricting the poor uses of AI. To me, ethics is all about responsible use. It’s beyond just rules and regulations. It is about a framework to cultivate the two pillars of responsibility, which are moral agency and accountability. By having these two aspects together, you can ensure that AI is there to do good above all else.”

“It obviously depends on the size and budget of the company in question. Smaller companies, for example, may not have the budget to implement their own versions of AI. But I very much doubt that there will be any companies left in five years that will not be using AI in some way, shape, or form. It will simply become a part of how we do our jobs.”

But what is the ultimate end goal for AI implementation? The end goal is basically a quest for excellence. It’s about improving long-term growth while setting the foundations for medium-term growth and long-term success. It’s about using it for deep thinking, which is critical for productivity, creativity and innovation.

The perception of AI has changed somewhat over the last few years. At first, people were sceptical about the technology, as is often the case with new and influential developments. People were sceptical about the repercussions of AI, but also whether or not it would actually live up to the hype. In fact, there is still an overconfidence about what AI can achieve on its own. 

AI models will become trained on inputs that have been created by AI itself. At that point, it is likely that human characteristics like creativity may be lacking, which may cause the quality of AI-produced outputs to diminish. This is why we need to be aware of becoming too overconfident about the abilities of AI. At some point, we might reach the ceiling in terms of its performance, and we must be mindful of that. 

If you have never worked with AI before, you should consider AI like a scientific experiment – it can go wrong. Of course, it can go right too, but it is important to ensure that human dignity and human responsibility are the guiding principles behind not only why we use the AI system, but also how we use AI.