by SAMEER PARKER
SAMSUNG has recently come under fire for being discovered using artificial intelligence to enhance images of the moon. It turned out that their phones were not improving their images at all, but rather completely faking them.
When you point your phone’s camera at the moon and zoomed in, the phone’s AI would overlay an existing library image of the moon, fooling you into thinking you had just shot a stunning photo of the moon.
However, the resulting photo is not one that you took; rather, it is a professionally captured moon image that has been digitally overlaid over the hazy moon image that your phone took. All done with AI, and most crucially, without your knowledge.
There are numerous reasons to be excited about the benefits that technology like AI can provide.
There are numerous reasons to proceed with caution during this stage of human progress. As with some new technologies, some warn of an impending hazard while others support and promote it without considering the risks.
Some may view the AI revolution to be no more significant than the introduction of the television or smartphone. But this is not the case for several reasons, particularly for Muslims who are cautioned not to become too immersed in this ephemeral world.
One of the fundamental differences is a notion known as suspension of disbelief. Suspension of disbelief is a state in which we believe something, even though our conscious minds know it is not real or possible. No adult of sound mind, for example, watches a Superman movie and believes that a man can fly. However, we allow ourselves to believe what we see on television for the duration of the film, without criticising or challenging the reality that we are watching a man fly. Our disbelief has, accordingly, been suspended or paused for 90 minutes. Importantly, when that time has passed, we become well aware that what we have witnessed was a work of fiction.
We also tend to be forgiving of ourselves for being immersed in the story and believing it while we were watching it.
Unlike television, which we are aware of viewing, AI will be interwoven into our lives in a much more subtle manner.
This means that we may be completely oblivious that we are experiencing anything generated or controlled by AI (as in the case of the Samsung moon). Generative AI is a sort of AI that can generate new material or content from current data or inputs.
There are existing tools that employ Generative AI to create text, images, audio, and video.
One of the most popular recent AI developments is ChatGPT, a natural language processing tool. It allows users to enter questions or prompts in a manner that mimics a normal human conversation, and it responds accordingly.
It can write speeches, perform complex calculations, explain difficult concepts, generate code used to create apps, and considerably more. Ask ChatGPT a question, and it will respond.
At its core is a massive amount of data that is accessed by a system that can comprehend, interpret, and exchange that data as useable information in unprecedented ways.
It can even generate responses in your preferred style. For example, you could ask it to explain how flowers grow in a writing style appropriate for a 6-year-old and another response appropriate for a middle-aged scientist.
Both responses will be vastly different. Other tools like DALL-E can create real-looking digital images after being given a prompt in natural language.
Another recent development in AI is a type of content called Deep fakes. These are fake images or videos created using Generative AI and designed to appear real.
AI tools can analyse references of a person speaking and, after receiving a prompt from the user, outputs a video of what appears to be that same person speaking.
To the untrained eye, it may easily appear and sound authentic, which brings up the critical point in the matter: Some of the most obvious ethical issues are found in creations such as deep fakes. The pessimist can argue that misusing Generative AI has a high potential for fitnah in the Muslim community.
If the dunyah (earthly abode) itself is an illusion, how do we prepare ourselves for an onslaught of what will inevitably become another layer of deception?
Fortunately, Muslims should be well-prepared to deal with this quandary. To begin, acknowledge that the dunyah is a delusion as Allah has advised in the Quran (Al-Hadid, 57:20): ‘What is the worldly life except the enjoyment of delusion.’
We are also guided by the Quran to hold truth dearly and very specifically, and not to mix fact and falsehood: ‘Do not mix truth with falsehood or hide the truth knowingly.’ (Al-Baqarah, 2:42).
In Surah Al-Hujurat (49: 6-8), we receive guidance on verifying information before acting on it. Unfortunately, the widespread dissemination of false information among Muslims during the most recent global epidemic demonstrated how far we still have to go in this respect.
The awareness and guidance provided by the Quran should be part of the Muslim approach to AI, both as contributors to the technology and as users of it.
This AI era provides us as Muslims with an excellent opportunity to reflect on our world and its delusions.
We should consider not only our own readiness as individuals but also the practical readiness of the Muslim community organisations on which we rely.
Are Islamic schools prepared to deal with AI-generated material submitted as original work by scholars? Are our judicial agencies prepared to rule on whether a new AI-generated translation of the Quran, no matter how technically correct, is permissible or even desirable?
What about the Shariah compliance of an investment tool that manages itself using AI? At least for now, we can all agree that no Samsung phones should be used for moonsighting, as it may complicate matters further.
Given that Muslim polymath al-Khwarizmi is widely recognised as the founder of algebra and the creator of algorithms, it would be nearly imprudent for modern Muslims to overlook AI.
One could even argue that Muslims should be actively participating in the development of AI tools.
Because these technologies are based on human data input and programming, Muslim contributions can ideally assist decrease bias and have a positive impact on the ethical frameworks and future of AI.
Professions such as Prompt Engineer, AI Developer, Data Scientist, Big Data Architect, Machine Learning Engineer, Data Analyst, Natural Language Processing Engineer, Researcher, and others, are all available in the field of AI.
Perhaps we should all take an interest in AI and other technologies, and find ways to encourage this interest in young Muslim minds at school level.
While artificial intelligence has the potential to revolutionise various industries and improve our daily lives, it also poses significant risks that must be closely monitored. As AI becomes more integrated into our society, it’s crucial that we ensure its development and deployment aligns with ethical and moral principles, and that it does not result in unintended harm or bias. By keeping a close eye on the advancements of AI and enforcing regulations, we can leverage the benefits of this technology while mitigating its potential negative consequences.
The last paragraph you just read was written entirely by ChatGPT.
- Sameer Parker is a social media and technology practitioner at Social Natives socialnatives.co.za.
This article was first published in the June 16, 2023 print edition of Muslim Views.