AI in Southeast Asia: Muse or menace? How artists in the region are grappling with new technologies

Advertisement

Asia

AI in Southeast Asia: Muse or menace? How artists in the region are grappling with new technologies

Artificial Intelligence (AI) is becoming a creative force both within and outside gallery walls. But legal uncertainties could hinder the hype. The latest in a CNA series on AI in Southeast Asia looks at how the technology is redesigning artistic practices, and what lies ahead.

New: You can now listen to articles. Sorry, the audio is unavailable right now. Please try again later.

This audio is AI-generated.

Amanda Oon

SINGAPORE: From his home office in Manila, Mr Patrick Cabral has built a city. 

The metropolis is traditional; the buildings’ beehive structures and nipa palm-thatched roofs are similar to Filipino indigenous bahay kubo houses. But it is also modern. The photo-realistic images show pre-colonial Philippines in a contemporary setting, which Mr Cabral created in 2023 using AI softwares Stable Diffusion and Midjourney. 

For the multi-disciplinary artist, who originally specialised in calligraphy, the latest technologies in artificial intelligence (AI) and machine learning have provided a new medium for his ideas, and also a platform to reimagine his country’s history. 

“(My) experiments aim to answer the question on how architecture and fashion would have evolved if we weren't colonised,” he told CNA, referring to the Philippines’ era under Spanish rule from 1565 to 1898, and various periods under American colonisation and Japanese occupation from 1898 to 1946.

Over recent years, Southeast Asia’s AI creative potential has not only benefited artists like Mr Cabral but also extended past the walls of galleries and studios. Machine learning tools have been used to improve livelihoods, boost medical progress and explore identity and heritage.

But even so, concerns in the artistic world over issues such as copyright and human replacement have caused some to question if the AI race is accelerating too fast.

THE START OF THE AI RACE

It might seem like a modern phenomenon, but the first use of AI as an artistic tool was in 1973, when British-born artist Harold Cohen debuted his computer programme AARON, able to generate basic black-and-white images. 

By 1979, AARON 's work had featured in an exhibition at the San Francisco Museum of Modern Art (SFMOMA). 

AARON’s algorithm was trained through lines of code that Mr Cohen designed and fed to a robotic arm to teach it to mimic the drawing process.

With the development of deep learning in the early 2000s, technologies developed which were able to not just replicate photos, but also generate their own, based on training data. These were produced at high speed and, unlike AARON, with no need for a robotic arm to do the work.

The first such text-to-image programme, AlignDRAW, was developed by then 19-year-old computer scientist Elman Mansimov in 2015.

Over the next decade, several new text-to-image generators followed. Unlike AARON’s algorithm, modern AI image generators mainly use text prompts, and train their algorithms by scraping data from swathes of existing images online.

For instance, DALL-E, introduced in 2021, was able to combine learnt concepts to depict abstract ideas, such as an avocado chair. Following in 2022, Midjourney uses AI to capture and replicate artists’ specific styles.

But for many artists in Southeast Asia, like Mr Cabral, their use of these programmes soon posed some cultural challenges.

“If you prompt (an AI algorithm) with ‘bamboo house,’ it will give you bamboo houses that are statistically on trend,” he told CNA. “But they don’t understand regional concepts, like what I mean by bamboo houses from pre-colonial Philippines.”

According to Mr Cabral, the lack of authentic depictions of Southeast Asia available on the internet meant that AI generators weren’t able to produce accurate depictions of regional clothes, people or architecture.

His experimentation highlighted the scarcity of representation in the visual arts and one of the key limits of these AI tools. Eventually, he was able to create lines of code he could feed directly into the programmes to train his AI to recreate Filipino motifs and landscapes.

He hopes his work can help shift the needle on a self-perpetuating cycle of AI artwork trained on western-centric sources and help regional artists reclaim their narratives.  

“Technological advancement holds particular promise for the regional art scene, especially for those of us from countries whose creative expressions are yet to gain global recognition,” he said.  

Indonesian artist Rimbawan Gerilya agrees that the potential of AI technologies to explore post-colonial cultures could help regional artists consider and challenge their own histories, and raise their profile. 

“They can use generative AI to reconstruct their identities, using their heritage as a reference instead of borrowing from dominant cultures to fill in the gaps,” he told CNA.

AI’S MOVE INTO MAINSTREAM

For Kuala Lumpur-based photographer Grey Chow, AI has helped open new commercial avenues.

His cheerful, colourful images of corgis smiling against psychedelic backgrounds, or a storm cloud spitting lightning over a cluttered desk, are all generated through text prompts, which he fed into Midjourney software. 

“Using AI, I’m able to generate more than a thousand of images monthly. This has opened up an additional revenue stream for me,” he told CNA.

The 39-year-old artist also runs AI image creation workshops, teaching participants how to choose the best prompts to optimise image creation and maintain a consistent style. The two-day workshop costs RM150 (US$31.77). 

“AI offers a significant advantage by lowering the barrier for art creation, making it more accessible and enjoyable for everyone,” he said. 

Galleries in the region are also capitalising on AI’s increasing accessibility. 

In September 2023, Singapore Art Museum launched ‘Proof of Personhood,’ a five-month exhibition exploring the interplay between art, identity and AI.

What struck curator Duncan Bass was not just “the number of people that have access (to AI) … and the speed at which they can produce this content,” but that it was also a talking point for those who don’t use it.

During the exhibition’s first weekend, Mr Bass led public tours of the gallery. 

“At the end of these tours, I was receiving questions from visitors in their 70s or 80s about the sentience of AI, whether or not we should consider these AI entities,” he told CNA. “For me, (that was) the sign of success.”

For Singaporean artist Charmaine Poh, whose work was exhibited in ‘Proof of Personhood,’ AI provided a platform to speak out.

‘GOOD MORNING YOUNG BODY,’ is a seven-minute video installation, starring a childlike deepfake reincarnation of the artist. It was created from footage of the young Ms Poh as a pre-teen actor, and her speech was trained off texts on media theory and cyberfeminism, a term referring to the interplay between gender, technology and the internet.

The avatar was able to speak out against the trolling and sexualisation Ms Poh had received as a 12-year-old television star.

“In this particular case, AI became a tool to retell a story from another perspective. The deepfake works as a skin, woven together by the data gleaned from footage of my 12-year-old body,” she told CNA.

“I think deepfakes can be interesting when they are used to question the narrator of a story, or to shift assumed positions of power. When it comes down to it, whose bodies do we regard as authoritative, and whose do we dismiss?”

Around the same time in the city-state, the ArtScience Museum was also turning its focus to AI.

Ms Deborah Lim, curator of the exhibition ‘Notes From the Ether: From NFTs to AI,’ which ran at the museum from August to September 2023, told CNA that one of the project’s key achievements was exposing AI art to new audiences, so that it could “resonate with not just those involved directly with technology and the web3 space, but the general public as well.”

For Ms Jo Ho, who is one half of Singaporean art collective jo+kapi, the process of AI art generation isn’t too far removed from the human process of creativity. It simply speeds things up.

“AI is trained on this huge data set that is based on our data, the things that we've produced, the things that we've created,” she said.

“(It) expresses something that you would have eventually gotten to in maybe 200 hours in or 2,000 hours. But AI is doing it at a much (more) rapid pace.”

Sometimes though, the tools can generate new perspectives. For one piece, Ms Ho remembers training an AI algorithm off datasets of animal skeletons. The programme was able to use the information to create new, fantastical, animal forms.

“I trained the dataset … but the machine created the new forms … it was interesting to have an element of surprise,” she told CNA.

Her collaborator, Mr Kapilan Naidu, has also experienced the algorithm’s unexpected results.

A prompt to capture the effects of Hurricane Ian, which hit Cuba and the US in 2022, produced a drone’s eye view instead of the on-the-ground perspective a human artist would have instinctively chosen. 

But he remains sceptical whether this is a sign of AI’s inherent creativity, at least for now. 

“Do androids dream of electric sheep? I think it's still early days for us to say that AI is being creative willfully. Ultimately if you were to pare down all the technology, it's still a statistical engine … making decisions based on preconceived notions of the data set it's been trained on,” he told CNA.

AI BEYOND THE ART WORLD

While artists consider the future, in the brightly-lit wards of Singapore’s Jurong Community Hospital, AI is looking to the past.

A team at the 400-bed institution launched an art therapy programme last year, using AI tools to help recreate formative events for dementia patients. Art therapist Lee Sze-Chin and his art collective Kronoscopes asked participants to recall happy moments from their lives and draw doodles from these occasions on small coasters. 

These were fed, along with information from patient-therapist conversations, into Midjourney, which transformed the prompts into detailed visual recreations of the memories. 

One participant in his 60s, who preferred to be known as Mr C, recalled helping his mother at the local rubber plantation as a nine-year-old child. He was moved to tears by how closely the AI-generated image resembled not only his memory, but also how much its main figure "somewhat looks like me."

“I was … surprised at how the resulting images generated quite visceral responses in some of the patients,” Mr Lee told CNA, recalling the project. 

Almost a year on, the art therapist is still enthusiastic about how AI can revolutionise reminiscence therapy, a form of treatment using memories to stimulate mental activity and improve well‐being.

He told CNA that patients gained a renewed sense of identity and self-worth and were able to process the emotions the AI artworks unearthed with their art therapist, whether those were positive or negative. 

Mr Lee added that AI technologies had broken down barriers previously holding back those whose motor skills or cognitive ability would restrict them from traditional techniques. 

Alzheimer's Disease International, the worldwide federation of Alzheimer’s and Dementia Associations, estimates that by 2025, 5.23 million people in Southeast Asia will suffer from Dementia. Reminiscence therapy can boost positive emotions and aid present relations, according to social service agency Dementia Singapore. 

But while Ms Buvenasvari Pragasam, an art therapist at Solace Art Psychotherapy based in Singapore, agrees with the beneficial breakthroughs AI can bring in recreating memories, she warns there are limits to what it can do for patients. 

“AI will not give clients the sensorial and kinesthetic experiences that they experience when interacting with physical art media, such as clay,” she said.

“AI makes it easier for anyone to bring their imagination to life. (But) in art therapy, it is important for the art therapist to observe the client actually making the artwork (without technological help). The finished artwork is secondary.” 

CONCERNS AND CHALLENGES OVER AI IN ART

Questions over the creative process and who is behind it have also raised concerns in the art sector.

Curator Ms Lim warned that “with the use of digital tools such as AI, it is increasingly difficult to attribute creativity or originality to just the artist alone”.

With many generative tools trained off the styles and works of existing artists, some in the industry are raising concerns over how creatives will be able to protect their identity and work. 

“I think the issue really stems from these services … training algorithms in (an) unethical manner … essentially just crawling the internet for artworks and training on specific styles,” said Mr Naidu.

“It's … allowed random people to wear the skins of artists, which I think is quite disturbing.”

Professor Risqi Saputra, assistant professor of data science at Monash University in Indonesia, agrees that the ease at which AI can replicate artists’ styles poses a threat to originality in the market.

“A notable example is the ease and speed with which even a mediocre artist can now apply the iconic style of Vincent Van Gogh's 'The Starry Night' to their own drawings using style transfer techniques,” he said.

In 2023, a group of USA-based artists launched the first class action lawsuit against several AI generators, including Midjourney, DeviantArt, and Stability AI accusing the companies of copyright infringement. 

The judge dismissed many of the artists’ claims, noting that two of the three artists had apparently never registered their disputed works with the Copyright Office. 

“Most countries are still trying to wrap their heads around … understanding and then regulating if, at all, AI,” said Mr Ronald Wong, Deputy Managing Director at Singaporean law firm Covenant Chambers. 

He pointed out that, although there is a requirement for human authorship under the Singapore Copyright Act, who that human author is – the person who inputs the prompt, the artist whose creation was used to train the algorithm, the owner of the AI programme or no-one at all – remains a legally grey area. And this could be intentional.

“In the process of reviewing and then revising the Copyright Act … the review committee …  deliberately left open the question of  computer-generated work (and who) owns the copyright,” said Mr Wong.

“I suppose it was sort of a wait and see kind of approach, let the courts decide if a case comes up before the court,” he added.

They could be waiting a while. Mr Wong thinks it’s unlikely that similar cases to the US lawsuit will reach Southeast Asia. One reason, he said, is the inherent power imbalance between international technology companies and independent regional artists. 

“I suspect a lot of this is driven essentially by who's got the deeper pockets … it’s hard to imagine someone like a Singaporean artist would have the resources to sue something like an OpenAI or Microsoft,” he said. “(And) a lot of these companies are not in Singapore (so outside local jurisdiction).”

LEANING INTO AN AI FUTURE

While the AI art bandwagon is in motion, some experts warn that it hasn’t accelerated at the pace the hype implies.

“The length of the video generated by Sora, one of the most prominent text-to-video models, still does not exceed one minute,” said Professor Bo An, Professor in Computer Science and Engineering and Co-Director of the Artificial Intelligence Research Institute at Singapore’s Nanyang Technological University.

“Developing new techniques to more accurately encode the prompts provided by users to generate desired outputs (such as) images that exactly match the user-provided prompts, and long-duration videos, would be important to apply AI models to the field of art.”

Professor Risqi takes a bleaker view of the future.

“If AI algorithms become too advanced, there is a risk that they could replace human artists altogether. This might lead to a loss of jobs and opportunities in the creative industry,” he warned.

But for Mr Cabral and his precolonial, virtual city, looking forward means looking back.

What is happening with AI in the region reminds him of a specifically Western period of history, an era of fervent cultural and artistic activity that marked the transition to modernity in the 15th and 16th centuries.

“The concept of creativity is undergoing a significant transformation,” he said. “I perceive a shift reminiscent of the Renaissance—a period devoid of rigid distinctions.

"In this new era, an individual can seamlessly embody the roles of a mathematician and a poet simultaneously.”

Source: CNA/ao(kk)

Related Topics

artificial intelligence technology arts

Advertisement

Also worth reading

Content is loading…

Advertisement

Expand to read the full story Get bite-sized news via a new
cards interface. Give it a try. Click here to return to FAST Tap here to return to FAST FAST