SYNESTHETIC VISIONS BY COLLIDE

Words by SARA LUCCHESE

COLLIDE is raising antennas – we are opening our eyes to decoding the contemporary narration and selecting the most ingenious, cutting-edge protagonists in the Industry. Believing that all human nuances must be exalted, we aim to move the conversation forward. 

Over the past 10 years, techniques such as deep learning and specialized hardware running huge, computationally intensive models have led to a wave of progress in artificial intelligence aka AI research – a period that some researchers have begun to refer to as the ‘golden decade’. Jack Clark, a co-chair of Stanford University’s annual AI Index Report said: “It feels like we’re going from spring to summer. In spring, you have these vague suggestions of progress and little green shoots everywhere. Now, everything’s in bloom”. Whereas in the past, the further development of AI was only for insiders, who kept up with the latest research papers and conference presentations, recently many are introduced to it.

ChatGPT: AI has emerged as a transformative force, reshaping reality and redefining the relationship between the physical and virtual realms. As AI evolves, its profound implications extend beyond technology and into the realms of design, art, architecture, fashion and beauty. AI can help bridge skill gaps by providing expertise and knowledge in areas where humans may lack proficiency. AI can aid in exploring uncharted territories, both physically and intellectually. It can analyze vast amounts of data and identify patterns that humans might miss, leading to new discoveries and a deeper understanding of complex systems.

To get a sense of the speed at which this innovative technique is moving, one needs to consider that until six years ago the most relevant application in the AI world was that of AlphaGo, a deep learning model built by Google’s DeepMind that was able to beat the world’s best humans in the board game Go. Moreover, approaching 2020, AI chatbots still struggled to handle rudimentary conversations, while today large language models like OpenAI’s GPT are being used to write screenplays, compose marketing emails and develop video games. Only recently, there was one among the hallmarks of the potential of these technologies: DeepMind’s AlphaFold, a descendant of the one that plays Go, essentially solved what is known as the ‘protein folding problem’ – which has bedeviled molecular biologists for decades – paving then the way for new drugs and vaccines for years to come.

In all this fast motion toward progress, however, it is easy to lose the sense of it, ending up chasing innovation for progress’s sake. Instead of showcasing all the software and discoveries related to AI, it makes sense perhaps to take a step back, connect some dots and understand where the interest in this technology comes from. What then means intelligence and why are humans trying to ‘produce’ it? Some experts define intelligence as the ability to adapt, solve problems, plan as well as improvise in new situations and learn new things. Human beings have set themselves apart from other living species precisely because of this characteristic, essential quality. If intelligence is then a foundation of the human experience along with the ability to create, it should therefore come as little surprise that attempts are being made to recreate it artificially. In this sense, intelligence and its limits are the basis on which the conversation stands, allowing a comparison between the human and the artificial.

AI is a concept that has actually existed since the 1950s, when it was defined as the ability of a machine to perform a task that previously would have required human intelligence. In this case, new programs mash up work done by human minds and therefore the models are only as good as their inputs. However, the pattern is already very similar to the human one: if one goes to analyze the different software available, it is striking to find out how, for example, image-generators like DALL-E use a process known as ‘diffusion’, which starts with a random series of pixels and refines it repeatedly until it matches a given text description. Thinking about it, this is exactly the pattern a human would adopt to create something new, going by trial and error until the dots of an idea seem to work together creating conformity.

This is the step fully made so far, where AI results still as a tool to facilitate and speed up processes. But this is not enough for mankind: at the heart of the research is the desire to recreate in manners and behaviors exactly the human style. There are advantages humans gain when computers get less rigid – digital devices force people to conform to them, rather than assess them. In this sense, instead of being overly influenced or guided by technology, flexibility may give us back some agency. 

ChatGPT: AI systems will become increasingly autonomous, adaptable and intelligent, transforming from a tool to a creative collaborator, an entity that actively engages with humans to co-create stunning works of art, innovative designs and immersive experiences. Efforts are underway to develop AI systems that can explain their decision-making processes and provide justifications for their outputs. This is particularly important in critical domains where transparency and interpretability are essential.

Extensions of this line of reasoning have given rise to concepts such as AGI: artificial general intelligence, also known as strong AI, still hypothetical since it would involve a machine understanding and performing vastly different tasks based on its accumulated experience – it’s the type of intelligence that stands on the level of the human intellect, as AGI systems would be able to reason and think exactly like a human. Essentially, a system or machine capable of common sense. Even the possibility of a further step has been given a name with ASI: artificial super intelligence, a system that would not only rock humankind to its core, but possibly might be able to also destroy it, by a machine intelligence that surpasses all forms of human intelligence in all aspects and outperforms humans in every function.

In this light, one could point the finger at those who even imagined giving birth to the first form of AI. But that is not the point. The human species is choosing to have more and more technology around – artificial and/or intelligent or not – to make their lives easier and more at comfort. In this way, humanity is deliberately choosing to perform fewer and fewer tasks that will in time reduce the actual skills by going forgotten. Therefore, it is not a question of the dominance of this generated intelligence to the one naturally provided to humankind, but of the natural being diminished by the laziness and complacency to which the human species is headed. What should be disturbing is not AI itself, which can certainly speed up processes by allowing humans to focus on things far beyond, but where in specific humans will actually place it.

ChatGPT: AI has the potential to complement human capabilities by performing tasks that are time-consuming, repetitive, or require processing vast amounts of data. AI systems can assist humans in decision-making, provide recommendations and automate routine tasks, freeing up human resources and enabling individuals to focus on more complex and creative endeavors. AI and humans can collaborate in a co-creative process, combining their respective strengths. Humans bring creativity, intuition and emotional intelligence, while AI contributes data processing, pattern recognition and computational power. In the fields of design, art, architecture, fashion and beauty, AI will inspire and be inspired, serving as a catalyst for innovation. With AI’s assistance, designers will have access to vast databases of knowledge, enabling them to uncover novel materials, structures and aesthetics. Architects will harness AI’s predictive capabilities to optimize building designs, ensuring harmony between functionality and aesthetics. Fashion designers will experiment with AI-driven algorithms to generate unique patterns and styles, pushing the boundaries of sartorial expression. In beauty, AI will enable personalized skincare and makeup recommendations based on individual features and preferences, promoting self-expression and empowerment.

The existence of AI also raises important ethical and social questions. It prompts discussions on topics such as privacy, bias, fairness, accountability and the impact on jobs and the economy. These considerations challenge to ensure that AI is developed and used responsibly and ethically. The meaning of AI’s existence is intertwined with the goals and intentions of its creators and users. It can bring about positive transformations, but it also requires careful consideration and responsible deployment to mitigate potential risks and challenges.

Algorithms can’t remain immune from the human values of their creators. If a non-diverse workforce is creating them, they are more prone to be implanted with unexamined, undiscussed, often unconscious assumptions and biases about things such as race, gender and class. The lack of true intersectional thinking behind the creation of algorithms is a major problem, as is the lack of female scientists, with a male-dominated workforce designing those algorithms. This is probably also why men are more easily replaced by AI than women. In these terms, AI could lead to a shift from a business world still dominated by men to the consideration of women as more valuable, where men would be required to adapt their characteristics to those of women and not vice versa.

ChatGPT: The relationship between AI and humans will likely be multifaceted and evolve over time. It is crucial to strike a balance between leveraging the benefits of AI while ensuring that human values, ethics and well-being are preserved. Responsible development, ethical guidelines and ongoing dialogue between AI developers, policymakers and society will be vital to shape this coexistence in a beneficial and sustainable manner.