Disasterchannel.co,- A feeling of awe arose when watching the film “The Imitation Game”. This film tells the story of a British mathematician named Alan Turing who developed a code-breaking machine called “The Bombe” for the British government. The Bombe was created to decipher the Enigma codes used by the German army in the Second World War.
Turing’s machine is considered to be the first functioning electro-mechanical computer, as well as one of the early births of Artificial Intelligence. Even the best mathematicians felt unable to do the tough task, but The Bombe managed to decipher the Enigma code. This made Turing wonder about such machine intelligence. In 1950, he published the landmark article “Computing Machines and Intelligence” in which he described how to make intelligent machines and in particular how to test their intelligence.[1]
Since then, AI has continued to develop and be widely used. Artificial intelligence (AI) has revolutionized many industries by performing tasks that typically require human intelligence to complete. AI contributes to complex scientific and engineering workflows by simulating, augmenting, or enhancing human intelligence in an efficient and precise manner.[2] Today, due to the rise of Big Data and increased computing power, AI has entered the business environment and public conversation.[1]
Unknowingly, our lives always intersect and are filled with the use of AI. If Alan Turing used to crack the Enigma code with The Bombe, now almost everyone uses AI to solve the problem. Currently, there are lots of problem solving practices that are solved by AI. As an example, if we want to know something, we will ask ChatGPT to find the answer. The use of AI in solving problems has spread to various fields, such as health, retail, manufacturing, banking, the public sector and other fields.
There is one problem that always has a big impact and must be faced by Indonesia. The problem was a catastrophic event. As a country located at the meeting point of the world’s plates, Indonesia has a multitude of disaster threats, ranging from earthquakes, volcanic eruptions, landslides, floods, tornadoes and so on.
Based on data from the dibi.bnpb.go.id page, throughout 2023, there were 863 disaster events. This disaster has resulted in 2417 houses being damaged and 4511 people being displaced.[3] So many disasters have befallen this country, creating complex problems that demand fast handling. Disaster is a cycle, in which there is a possibility that the event will be repeated in the future. AI has great potential to be one of the keys to reducing the impact of future disasters by reducing existing disaster risks in an area.
Currently, AI has been used in various fields, including disaster management. Artificial intelligence, especially machine learning (ML), is playing an increasingly important role in disaster risk reduction, from forecasting extreme events and developing hazard maps to real-time event detection. For example, in Georgia, the United Nations Development Program (UNDP) created a nationwide multi-hazard early warning system (MHEWS) to help reduce the exposure of communities, livelihoods and infrastructure to the threat of weather- and climate-driven disasters. Another example, AI is used in geodesy to detect tsunamis. The application of real-time Global Navigation Satellite System (GNSS) processing for ionospheric positioning and imaging provides a very significant improvement in early warning of tsunami disasters. [4]
The development of the use of AI in disaster management is not as smooth as expected. There are many problems and challenges that must be faced. Research by Sun et al, (2020) suggests that many challenges in implementing AI in disaster management are caused by data-related issues, such as accessibility, completeness, security, privacy and ethical issues.
Making accurate predictions with AI techniques requires large amounts of good data to build models. This data is not always available, if it is available it is incomplete, apart from that the data in a disaster environment changes dynamically. On the other hand, there are non-reproducibility challenges, because disasters occur irregularly with varying impacts in various regions. The fact that, as the threat of disasters and society continues to evolve, it is possible that this may change the usefulness of the attributes used to train the initial model. Not to mention, there are many areas that have not been touched by the application of various AI methods.[5]
Disaster management must be carried out holistically and inclusively. If AI is used in disaster management, then its use must also be inclusive. In his writing, Kevin Blanchard reveals that the integration of AI in advancing disaster risk reduction practices has inherent challenges, especially in ensuring the representation of marginalized groups. An interdisciplinary approach is needed, bringing together insights from fields such as computer science, social sciences and ethics. Collaborations like this not only help in overcoming the challenges posed by AI in DRR.[6]
AI is used as one of the offered disaster management solutions in this country. The challenges of its use must be faced by realizing collaboration. What’s more, the distribution of Indonesia’s territory is so wide with the diversity of tribes, cultures and also the high inequality of technological progress. It is important for all of us to ensure that any group is not marginalized in the practice of using AI for disaster risk reduction efforts.
Author: Lien Sururoh
Translator: Nugrah Aryatama
Source:
[1] Haenlein, M., & Kaplan, A. (2019). A brief history of artificial intelligence: On the past, present, and future of artificial intelligence. California management review, 61(4), 5-14.
[2] Muthukrishnan, N., Maleki, F., Ovens, K., Reinhold, C., Forghani, B., & Forghani, R. (2020). Brief history of artificial intelligence. Neuroimaging Clinics, 30(4), 393-399.
[3] https://dibi.bnpb.go.id/kwaktu2
[5] Sun, W., Bocchini, P., & Davison, B. D. (2020). Applications of artificial intelligence for disaster management. Natural Hazards, 103(3), 2631-2689.[6] https://irp.cdn-website.com/cde3424c/files/uploaded/AI%20report%20-%20final.pdf
Photo: sah.co.id
