Disclaimer: This article comprises snippets from interviews with organizers of a few leading design competitions. The views in this article belong solely to the writer and aren’t representative of the design competitions themselves. All snippets are inserted in “quotes” for your convenience.
On the 16th of April this year, the obvious happened. By ‘obvious’, I mean a controversy surrounding the use of AI tools, which have seen a dramatic rise in popularity since last year. Boris Eldagsen, a German photographer, admitted to using an AI-based image generator to create an image that he then enrolled for the Sony World Photography Awards. The controversial part is that he won the award, having somehow skirted scrutiny by Sony and even its esteemed jury panel. On his website, Eldagsen issued a statement refusing to accept the award, saying that he “applied as a cheeky monkey” to find out if competitions would be prepared for AI images to enter. “They are not,” he added.
The Award’s response was swift, but not swift enough. Following Eldagsen’s admission, they immediately took back his award and removed the entry from their website… but what’s truly baffling is that these AI tools have existed for over a year, and have been a hot topic of discussion for the better part of that year… so how did the entire awards program not prepare for this? The Sony World Photography Awards just announced that they’re open for their 2024 edition, and the website simply lists that “Computer generated content cannot be the origin of the Entry”, with little mention of how they’re going to distinguish human-captured photos from AI-generated ones, especially given how stunningly realistic AI text-to-image models have got in the last few months.
This event sparked outrage within the photography community, and one can only assume that the Sony World Photography Awards’ credibility took a sizeable hit. It also got me thinking about whether design awards were vulnerable to this approach too. Rather than simply waiting for another Eldagsen-like moment, we reached out to multiple leading global design awards to ask them what their thoughts were, if they were concerned about AI-based entries, and whether they were prepared for the eventuality. Only three design awards responded…
Boris Eldagsen and his Award-winning AI-Generated Photo
The Legality of AI-based Creations
Before we really dive into how awards perceive AI-based design creations, it’s important to understand how the law views them. After all, most award rules and regulations state that you need to “own” what you create… and ownership with AI-based tools can be a pretty murky subject. As far as AI tools go, most leading companies like MidJourney, Stable Diffusion, and OpenAI grant you ownership of what you create, but the law works differently. At least in the US, the United States Patent and Trademark Office doesn’t grant copyrights to AI-generated work. If it isn’t created by a human, it isn’t eligible for copyright protection (as is the case with China). In order for an AI-generated work to be eligible for copyright, it needs to have a significant amount of human intervention or input in it. Laws in the UK and EU focus on “originality”, which is highly debatable too, given that nothing an AI creates is ever original – it simply samples work from large datasets. Additionally, in order for a work to be sufficiently original, it should be the result of an author’s “labour, skill or effort”, which can further muddy the water. Rules in countries like India are a little more accommodating, with multiple artists having copyrighted their AI-generated creations.
What do Design Awards have to say?
While design awards, in practice, aren’t too different from photography awards, the primary distinction lies in the fact that Photography is an art form, whereas Industrial Design is a human-centric process that’s an amalgamation of art, science, emotion, marketing, and consumer feedback… and that’s just not something AI can overwhelmingly replace overnight. The Design Intelligence Awards calls this the “long design chain”, versus something like Logo Design or Visual Design which is more of a “short design chain.
Of the three design awards that got back to us with responses, all of them have unanimously expressed that for now, design competitions and awards have nothing to worry about when it comes to the use of AI… primarily because AI has already played a pretty significant role in the submissions in the past. Uwe Cremering, CEO of the iF Design Award, highlights how AI has played a significant role as a feature in multiple products. “Let’s talk about the Fridge, for example,” he says. “The fridge is not a hardware product anymore… It comes with an app, with algorithms, with software, with artificial intelligence that reads your needs.” By just loosely defining AI, any smart product has a measurable amount of artificial intelligence in it, thus making the submission an AI-based submission. A regular phone just can’t compare to a smartphone with its myriad of features – does that make the smartphone an unfair participant? Not in the eyes of Cremering and iF Design Awards. He does, however, highlight that just the same way a fridge isn’t a hardware product anymore, the award’s jury can’t simply comprise hardware designers anymore either. The iF Design Award is increasingly looking at a diverse group of jury panelists who can effectively judge these entries. The Design Intelligence Award, the largest international academic award in industrial design established by the China Academy of Art in 2015, is taking things a step further by developing AI to assist its jury members.
An AI-Powered Jury… Imagined By Midjourney
The DIA AI Jury is just one example of how the international award program is looking at the future. “The disruptive changes in design tools and technical logic brought by AI force us to rethink the unique value of human designers,” say the organizers. To augment the abilities of designers all across the professional spectrum, DIA has also built a DIA BOT, which they describe as an “AI dialogue robot based on design vertical field data, helping designers at different stages to find industry orientation, professional mentors, design knowledge and cooperation opportunities.”
While awards are adapting to this new ‘explosion’ of AI tools, there’s still a consensus that at least for the foreseeable future, Industrial Design awards in particular have nothing to worry about. Cremering does highlight that they’ve included an entire section of AI in the iF Design Award’s trend report, and that for maybe the next 2-5 years, there’s really no danger of AI submissions completely replacing human submissions. For Prof. Dr. Peter Zec, founder/initiator of the Red Dot Design Award, AI just simply can’t be as nuanced as a talented, problem-solving industrial designer… and even if it can, that ultimately bodes well for both the design industry as well as the consumer that benefits from the award-winning design. We managed to catch Dr. Zec just as he was about to leave for Singapore to preside over the jury phase of the Red Dot Award: Design Concept, and have a 30-minute-long conversation on whether design awards need to ‘worry’ about AI the way photography awards do. His answer was incredibly simple – the award’s sole purpose is to look at the result. If it’s a good design, regardless of where it came from, it’s a good design. In a way it echoes what Uwe Cremering mentioned too, about how AI can help make products better… but for Dr. Zec, what matters really at the end of the day is “the result”. For the Red Dot Award: Product Design, the result is almost always a physical product, which an AI can’t make just yet; but even if it did, it doesn’t change any award program’s ethos of rewarding good, consumer-friendly, utilitarian design. Dr. Zec raises the example of companies using AI to create more efficient, aerodynamic vehicles and vehicle parts. If an AI were to help designers determine the best, most efficient shape for the wing of an aircraft, it’s definitely worth an award since it still satisfies the criteria of being great product design. “So this makes a big difference because we are not looking for the most creative designer, we’re looking for the best solutions and then I think every tool is acceptable to reach the best solution,” Dr. Zec concludes.
We asked AI to imagine a beautiful Design Award Trophy…
How should you proceed if you want to participate?
Even with this level of acceptance from award programs, it’s important to read between the fine lines before applying. Most awards strictly mention that applicants must own the rights to the work they submit – an area that’s fraught with complications. Most countries don’t allow you to copyright AI-generated work (as mentioned before); and more importantly, even if you were able to copyright AI-generated work, it’s still plagiarizing content by other creators. Unlike humans, who draw inspiration from things they see/experience and create original works of art, AI doesn’t work that way. AI works by sampling other works, creating a collage of borrowed bits from a massive database… and when it doesn’t hide its tracks really well, it’s easy to make an argument on the grounds of plagiarism (a lot of Midjourney users noticed the Getty Images watermark appearing on their photo generations because the AI had sampled the entire Getty database).
If you’re using an AI program that relies on a large dataset or a large language model (LLM), it’s best to err on the side of caution and not make AI-based content a part of your “submission”. These AI tools are an incredible source of inspiration and are great for brainstorming, but that’s all they should be. Maybe don’t use ChatGPT unsupervised when you’re writing your application to an award, or Midjourney images as a part of your submission… but that’s just my take. These tools can really empower designers and humans in general, but they can’t replace what we do or parts of what we do entirely. Moreover, read your award’s rules and regulations very carefully before applying for them, just to make sure your entry isn’t subject to extra scrutiny or worse… disqualification.
The Best and Worst Case Scenarios
Tim Berners-Lee, who’s credited as the inventor of the World Wide Web didn’t know how powerful it would become as a force for both good as well as for evil. Zuckerberg, when he built Facebook, didn’t realize his benign social network would be used as a massive force for misinformation and a tool to destabilize countries, cause genocides, and sway elections. It’s entirely possible that the people building these AI tools are oblivious to what the future could look like, so it seems like a fairly healthy practice to just look at both the best and worst-case scenarios.
The best-case scenarios are pretty evident. More and more designers will find the AI very helpful in ideating, visualizing, and building out concepts. Designers using these tools will be able to achieve great heights without needing large teams with diverse skill sets, helping equalize the divide between designers in different countries, with different levels of experience, and with different backgrounds and upbringings. It all sounds great on paper, but perceptions change the minute a debacle like the one with the Sony World Photography Awards occurs.
As someone who’s commented on new technologies for almost 9 years at this point, the worst-case scenarios sound a lot more dystopian. After all, the worst-case scenarios for humanity should definitely apply to all parts of humanity too – including awards. Every AI relies on a massive database, and let’s just say from experience that massive databases will always contain information that shouldn’t be public. Facebook’s database contains all your thoughts, moments, liking/commenting history, etc. Google’s database is a little more high-stakes, it has all your emails, your location, your bank information, and your biometric data if you use a Google phone. The issue with LLMs and large databases is that data security becomes difficult when the AI kind of has a mind of its own – someone prompt-engineered ChatGPT to give them Windows 10 registration codes… and it did. These LLMs and databases are breeding grounds for infringement/plagiarism that reflects poorly on everyone, and even malicious tools – an engineer got ChatGPT to write malware and it managed to write a working piece of dangerous malware.
ChatGPT can apparently generate software keys for free…
The AI can now do things that weren’t possible – it could analyze every winning design from a particular award program/competition to generate a strategy for what project is most likely to win an award. When asked about this, Dr. Zec expressed concern but also highlighted how this exact process of pattern-recognition could have been done 10 years ago without current AI tools too. Ultimately, for Dr. Zec, if the jury deems a design award-worthy and good for mankind, that’s really all that matters. On being asked whether design awards needed to create an overarching body or consortium to address and tackle these issues, Dr. Zec agreed that this would probably be required in the near future.
What the future will look like in an AI-driven world [Opinion]
The printing press was met with a lot of backlash from religious leaders (and their followers) who felt the technology could be used to spread heretical (unorthodox) ideas easily to the masses. The television was dubbed an ‘idiot box’, and people said it would ruin the brains of children because they would stop reading. The internet to this day is still met with concern because of how its presence threatens a sense of ‘control’. Control over how life should be lived, how things should be done, and how societies should behave. This is exactly the same opinion a lot of people have with AI too – that it evokes a feeling of not being in control anymore. However, the printing press was invented in the 1400s and it didn’t destroy humanity – humanity simply evolved to integrate it into itself as a means for progressing forward. Humanity will do the same with AI too. Honestly, I’ll be surprised if AI destroys us BEFORE climate change or war.
How awards evolve to deal with AI is a whole new discussion, however. It’s interesting to see the take of major design awards on AI submissions – especially with awards like DIA highlighting the threat to ‘short design chain’ categories versus ‘long design chain’ categories. Not one single design award we spoke to was in favor of ‘blacklisting’ projects created using AI, which shows that these awards are definitely curious to see how AI changes/elevates/augments human creativity… and that’s a good thing!
An Architect building Futuristic Societies
Images via Midjourney
The post How Are Design Competitions and Awards Planning On Dealing with AI-Based Submissions? first appeared on Yanko Design.
via https://ift.tt/WCsUkQg
Post a Comment
Note: Only a member of this blog may post a comment.