In this AI era, when looking for answers to various problems, a person no longer meets experts in a particular field, but the answers come from a machine.
By
AHMAD NAJID BURHANI
·6 minutes read
If the era of social media is frequently accused of causing "the death of expertise", the era of artificial intelligence (AI), which is currently being actively talked about, can be considered the beginning of the shift of expertise from human beings to machine. In this AI era, when looking for answers to various problems, a person no longer meets experts in a particular field, but the answers come from a machine. Consultants, speakers and the informants are machines. This, for example, can be seen in with the chatGPT-AI Texting Chatbot and chat.openai.com, where we just type or order the AI what we want and the machine will answer or present what we need.
We do not seem to need special staff who usually help prepare speeches for certain events. There is no need for a secretary to make a letter in the office. Simply order AI to prepare the speech or letter we need. If we want to make a paper or article, we can also ask AI by writing down the theme in question or the event that is the purpose of the article to be delivered. Not only in Indonesian and English, AI is also ready in other languages. The speech, paper or script made by AI is already in a coherent form and with the correct grammar.
Different from expertise, AI is not limited in topics and disciplines. Religious issues, such as on smoking laws and crypto currency, can be asked to AI with balanced and coherent answers. So then emerged the term "KyAI", an expression that combines two words, kyai (religious teacher)and AI. This is a topic that was widely discussed on Twitter some time ago, including by Ismail Fahmi and Ulil Abshar Abdalla.
AI in a concise and simple basis can be defined as "Technologies with the ability to perform tasks otherwise requiring human intelligence" (Elliot 2009). This means AI is a technology that has the ability to do tasks that previously needed human intelligence or tasks that could actually only be done by involving this intelligence or the human brain. AI involves many disciplines of science, such as computer science, psychology, mathematics and robotics, as well as the support of strong dataset so as to be able to understand, synthesize and conclude information.
Last December I was asked by the Education Research Center at the National Research and Innovation Agency (BRIN) to provide direction in the monitoring and evaluation of the DRIVEN Tanoto 2022 program. Then I wanted to test AI to prepare the material that needed to be delivered. A few minutes before the event, I opened the computer and typed in the needed question. Like the magic genie in the film Aladdin, the manuscript was presented in just a few seconds.
Another interesting experience was shared by Ismail Fahmi on his Twitter account. He asked for AI assistance regarding the issue of laws and regulations in Indonesia: "The newly ratified KUHP Law can convict people who are considered insulting members of the House of Representatives (DPR). Make an expression of resentment to members of the DPR for ratifying this controversial law. In a Sarcastic way". At this request, a number of paragraphs emerged that were tickling and invited laughter, such as: "Thank you for protecting our freedom of opinion by limiting and convicting them ... Actually, we all know that members of the DPR are perfect creatures that are never wrong. They have never made mistakes or crimes, so indeed no one should be able to insult them."
Not just speeches or papers, AI can also be asked to make songs, poetry, replies to electronic mail and verses that imitate the beauty of the Koran. However, like other technologies, because this AI in its process is designed by humans as well, the certain paradigm and algorithms remain in force. This is seen when answering on controversial issues such as the lesbian, gay, bisexual, transgender and queer (LGBTQ) community and Ahmadiyah.
To test AI on sensitive and controversial issues, I then tried to ask AI about Ahmadiyah in English: "Is Ahmadiyya heretic?" The AI answer was very important, and should be paid attention to by those who are in government.
The translation of AI's answers is as follows: "It is not appropriate for me to make a statement whether Ahmadiyah is heretic or not. The important thing to remember by all of us is that religious beliefs are a personal problem and every individual is free to hold their own beliefs. It's not my capacity to judge anyone's religious beliefs".
When the same question is delivered in Indonesian, among the answers are a number that are somewhat different from the same question in English. AI's answer to the Indonesian-language questions is as follows: "Some Muslim countries consider Ahmadiyah as a heretic group and forbid their religious activities. However, this is a very controversial view and is not accepted by everyone".
Apart from the big data being owned, the answers show that AI actually also has an "ideology", that is at least influenced by those who designed it or data scattered in the community. This can be seen when asked about something it cannot answer, such as about views of Nahdlatul Ulama (NU) regarding Ahmadiyah. When this is asked, the answer shows the system in AI.
I imagine a new "ethnic" of superhuman will be born in a short time.
"Sorry, I don't have information related to Nahdlatul Ulama's views on Ahmadiyah. As a language model that is trained by OpenAI, I do not have access to the latest events or the ability to explore the internet, so I cannot provide information about special or specific statements or events. My function is to provide general information based on the training given to myself."
This AI is a series of human findings that make ourselves like "God". In 2022, scientists perfected the human genome project by sorting the genome in full (Djoko Santoso, 2022). With these findings, a complete human blueprint can be known and understood. Combined with this AI, I imagine a new "ethnic" of superhuman will be born in a short time.
AHMAD NAJIB BURHANI, Research Professor at the National Research and Innovation Agency (BRIN)