Ben Meisner is the founder of the leading online photo editing platform Ribbet. com.
A wave of AI-powered photo and video editing tools has emerged in recent times, ranging from tools that simply save time by automatically removing backgrounds, to more advanced tools that replace humans (deep fakes) or generate completely new and realistic scenes from scratch, such as DALL·E and Stable Diffusion.
What is the driving force behind the creation of these tools and algorithms?
When you look at the development of AI technology in self-driving cars, there are obvious issues that are being addressed. For example, there is the potential benefit of fewer accidents, where approx 1.3 million people are currently killed each year as a result of road deaths.
But when it comes to photo and video manipulation, what are we ultimately working to solve? Are engineers creating these algorithms purely because new technology has unlocked the potential to create them, or is there a deep problem humanity is trying to solve?
This topic is highly debated within the creative ecosystem, with some companies going so far as to ban the use of AI-generated media because they could violate copyright law.
The conversation shouldn’t stop here though – beyond the question of whether AI-generated media are truly original, I think we should consider whether they work can be considered “real art”.
What impact does this have on companies and creative professionals?
This consideration has a major impact on business. If AI can elevate to activities traditionally considered human, then that will change consumer perceptions when it comes to future products that will similarly rise to those areas: think self-driving cars, AI copywriters and photographers or AI- personal assistants.
By this I mean that AI needs to be widely accepted as reliable and human-like before it can be trusted to do the work that humans have traditionally done. For example, before we trust a car to drive us around using AI, we need to have faith in AI in general. So if AI could generate something that was considered “real art,” it could be a step towards doing something that traditionally only humans could do.
Today, art is still generally defined as an activity performed by people, created by those who are trained to capture the essence of things and their feelings in an appealing form involving the imagination. We often think of art as those forms of expression that come from someone emotions and what we relate to on a human level.
But what happens when a new work is generated by an emotionless machine? Traditionally, we would consider that a callous act of referral. A silicon chip running a program and using millions of reference media to create something new – could that ever be considered an ‘artist’? It is certainly not an intentional art; it represents the intention of the user of the machine.
The issue of it being a reference work is not necessarily a problem for it to be considered true art. Great writers, musicians and painters have always been inspired by the work of their predecessors. Besides inspiration, there are rules – or the structured breaking of rules – that make or break a great work of art, and that can be taught, even to a machine. These rules exist everywhere – in our language, art, storytelling and business.
A computer can record these rules and structures and generate music, for example. However, could it ever show vulnerability? If AI were to engage in creative endeavors that stir our emotions and become widely accepted as authentic art, perhaps there would be one less thing that makes us unique as human beings.
It would be easy to stop here and declare AI-generated art as inauthentic. Photography has also fought a long and controversial fight to be considered true art. Today we’ve reached a point where it’s accepted, largely because there’s a human artist framing it. Could machine-generated art, which runs through an algorithm created by a human being and begins with a human inputting a sentence, ever be considered “framed” in that same sentence?
There would be benefits if it did.
Technology can enhance professional skills and capabilities, not simply replace them.
AI-generated art could give those people without traditional artistic abilities the opportunity to express their feelings and vulnerability through art. There would also be benefits for businesses. Today, people without any design skills can already generate a layout or design by using ready-made templates for whatever they are trying to create.
Imagine if at some point you could demand from a computer: “I need a new header for my website – something inspiring, with an image of a woman wearing our company’s shoes, in blue. Put that image in a collage of women with different looks, and give me 10 variations of that.”
AI-generated art can facilitate design and creative processes, reducing resources needed to launch a beautiful campaign visual, for example, and acting as a resource that the creative industry uses, not a tool that replaces it.
Be aware of copyright and licensing challenges.
At the same time, there are still discussions about copyright issues and licensing, which entrepreneurs should not ignore. The reference material used to train AI algorithms often contains copyrighted material, so accidental infringement is a possibility. The technology is not yet mature, so licensing should be taken seriously now.
As attractive as AI-generated art may be, care must be taken in how these tools are applied. It would be problematic to use it for an advertising campaign, for example, if copyright were violated.
At the moment, however, these AI generation tools appear to have limited real-world application. We seek a useful purpose for them rather than initially setting out to solve a particular problem. At the same time, however, I can’t help but think that this could all change as the technology matures and new and exciting potential applications emerge. I’m also drawn to the possibility that our definition of real art and what it encompasses could soon evolve into AI-generated art.
Janice has been with businesskinda for 5 years, writing copy for client websites, blog posts, EDMs and other mediums to engage readers and encourage action. By collaborating with clients, our SEO manager and the wider businesskinda team, Janice seeks to understand an audience before creating memorable, persuasive copy.