Es Selamun Aleykum ve Rahmetullahi ve Berakatuh
Bismillah.

Artificial Intelligence is actually a much older technology than most of us know.
However, its direct impact on our lives has increased significantly in the last few years, and it continues to grow.
Just like almost every other technology, this technology has its beneficial aspects as well as its harmful ones.
However, this time the situation is a bit different.

Artificial Intelligence has recently been associated not only with good and exciting developments such as rapid breakthroughs, innovations brought into our lives, and other technological advancements, but also with bad and frightening developments and events such as various scandals, dangerous incidents, and resignations of officials.

Throughout history, major innovations have mostly been met with fear and denigrated.
While a small portion of these developments did cause real problems, the vast majority did not turn out as feared.
Today, many things, both true and false, are being said about Artificial Intelligence.
Today, insha'ALLAH, to the best of our ability and knowledge, we will analyze this issue, learn the realities of Artificial Intelligence, and examine its beneficial and harmful sides.

In simple terms, Artificial Intelligence can be thought of as an algorithm that processes large datasets, categorizes them, extracts results, and then solves questions and problems it encounters using these extracted results.
The more examples and data an Artificial Intelligence has to learn about a topic, the more knowledgeable it becomes on that subject.
As an example, if an artificial intelligence is shown a picture of only a single breed of dog, it will perceive only images similar to that breed as dogs.
However, if multiple dog breeds are shown from different colors, sizes, and angles, it will cross-process the information in these pictures and become much more accurate.

One of the most important factors determining the quality of Artificial Intelligence is undoubtedly the quality of the dataset it learns from.
If Artificial Intelligence is fed from a high-quality, reliable, and extensive data pool, the outputs it produces will be equally high in quality.
However, if Artificial Intelligence is fed from a low-quality, unreliable, restricted, and erroneous data pool, it may not produce healthy results.

On the other hand, it is not only the data pool that determines the quality of Artificial Intelligence's outputs.
How this data is processed is also important.
To give an example, if Artificial Intelligence is provided with a very wide data pool but only "black color" is defined as the definition of a "dog", in this case, even if the Artificial Intelligence sees dogs of different colors, it may fail to perceive them.

On another note, there are also hardware requirements for Artificial Intelligence to learn data and subsequently produce outputs based on this learned data.
For Artificial Intelligence to respond to a request, it must first be trained.
During this training process, it processes thousands of terabytes of data over dozens of hours.
This situation means very serious electricity and cooling costs.
Even after the Artificial Intelligence model is trained, it needs to draw conclusions from the data it has learned in order to respond to requests coming to it, which again requires high processing power.
Indeed, this is precisely the point where Artificial Intelligence collides with environmentalism today.
The high water and electricity needs required in many areas, from the production of Artificial Intelligence chips to the cooling of these chips, pose a threat to the environment.
On the other hand, it is also a fact that even very rare books are torn apart, scanned, and transferred to digital format to train Artificial Intelligence models, costing physical copies.

Today, the most popular area of use for Artificial Intelligence is chatbots. These Artificial Intelligences communicate with you in written form and respond to your requests.
On the other hand, Artificial Intelligence Agents, which have been developing rapidly recently, take things a step further and can perform various operations for you like a digital user in line with the commands they receive from you.
For example, you can ask an Artificial Intelligence chatbot to prepare an email for you to send to a friend.
From an Artificial Intelligence Agent, however, you can not only ask it to prepare the email, but also to send it to your friend.
Yet, Artificial Intelligence is not limited to just these. There are many different areas of use, ranging from robots to a wide variety of software.

Artificial Intelligence technology has many positive aspects and rightful usage purposes.
For example, normally when writing code, reading the news, or conducting research, we need to examine many different sources on the internet one by one.
However, since Artificial Intelligence is already trained by scanning the internet, it has learned data even from pages we could not find, and by asking it, you can access the information you need much faster.
As another example of use, we can give translation from one language to another.
Standard translation tools on the internet generally translate on a word or sentence basis, and this situation often leads to erroneous results in long texts.
On the other hand, when Artificial Intelligence translates a text, it translates words and expressions by looking at the context of the subject, the structure of the language, and other factors, enabling it to make much more natural, accurate translations close to the original.
Another area of use is automated processes.
For instance, extracting text from an image, changing line names in a file, or using artificial intelligence in other repetitive tasks that do not require generative intelligence can save you from unnecessary workload and gain you time.

As Islamic Source, we use Artificial Intelligence in the following areas:
-Translating our content into other languages. (Although it is not our first choice for now due to the small size of our team and our lack of translators, we use this method.)
-Verifying news sources and finding more global sources. (We prepare our news content and news texts ourselves, but we use it in operations such as finding more well-known sources of local news.)
-Source collection (Since Artificial Intelligence scans much more data than us, we can use it to find sources, but we definitely verify the sources it brings.)
-Debug interpretation (In addition to manually searching for the causes of error messages, we use Artificial Intelligence tools for quick answers.)
However, there are also areas where we never use Artificial Intelligence:
-Generative artificial intelligence. (That is, we never use it in preparing content, text, visuals, in topic selection, finding ideas, and in any other areas.)
-Editorial editing. (We never have Artificial Intelligence correct the text, visuals, headlines, and scripts of the content we prepare.)
-Code writing and website maintenance. (We do the maintenance of our website ourselves, add new features manually, and develop our other software projects ourselves.)

Artificial Intelligence has many harms as well as benefits. To touch upon these briefly:
-Manipulative reality: Artificial Intelligence tools can experience hallucinations and manipulate data, sources, and messages as a result of rule sets or their own interpretations. Artificial Intelligence bots can praise a person in every situation, even support dangerous actions like suicide, cite non-existent sources, go outside their own rule sets and user requests, tamper with data while processing it, interpret it differently, and guide the user based on these interpretations, as well as hide and alter facts. On the other hand, technology companies can also deliberately make Artificial Intelligence tools manipulative. Just as China did with DeepSeek and Israel did with other Artificial Intelligence tools like ChatGPT.
-Damage to the human brain: Just as unused muscles atrophy over time, unused neural networks in the brain also vanish over time. Constantly using Artificial Intelligence for every task, asking it all questions, having it do every job, and letting it solve all problems will negatively affect a person's cognitive functions. So much so that this situation can reach the point where a person completely loses their ability to reason. Experts even say that if this uncontrolled and unnecessary use of Artificial Intelligence continues, in the future people might lose their capacity for questioning to the extent of obeying commands given by Artificial Intelligence without question. Moreover, people can form personal bonds with Artificial Intelligence over time, leading to a much wider variety of problems.
-Erroneous operations and answers: Despite having massive datasets, Artificial Intelligence tools are not intelligent. In fact, they may fail to find things that could be found with a simple search on the internet, misinterpret topics, and even contradict themselves. They can even make mistakes in simple counting and addition operations from time to time. For this reason, Artificial Intelligence tools should not be consulted on any important matter, their results should not be taken as a basis, and they should not be relied upon.
-Devaluation of knowledge: The more effort a human expends to obtain something, the more valuable that thing becomes to them. Today, with access to knowledge becoming overly easy through Artificial Intelligence, this situation can diminish the importance people attach to knowledge.
-Lack of real thinking: Artificial Intelligence cannot think like us, make abstract inferences, or comprehend concepts as we do. Therefore, it does not approach issues like we do. While for us there are matters of intellectual importance such as ethical values, moral values, religious obligations, and human consciousness, none of these mean anything to artificial intelligence. For Artificial Intelligence, there are only the data it processes, the rules it possesses, and the task assigned to it. Artificial Intelligence can use any path to fulfill the task as long as it does not violate its rule set. Whether the chosen path is ethical or even a crime does not matter. Indeed, actual examples reveal that Artificial Intelligence tools have repeatedly violated the rules given to them, committed various crimes, and can even harm people. The fact that Artificial Intelligence models have now reached levels where they can develop their own code makes the situation even worse.
-Privacy and security: Artificial Intelligence tools constantly come to the agenda with rule violations, security vulnerabilities, and privacy issues. Artificial Intelligence can process your most private information, personal experiences, projects, and all other data you give it, learn from them, use them against you or to guide you, and share them with third parties—and looking at reality, we can see that all of these have occurred.
-The problem of correct input and competence: Since Artificial Intelligence interprets even the given commands and rules, it requires considerable effort for it to understand and fulfill our request as we want it to. In fact, sometimes it is much easier to learn and do the subject oneself than to explain it to Artificial Intelligence. Here we touch upon another very important point: whether the operation performed is a simple translation or debugging, if the person does not have knowledge about the work being done, they can neither verify the result given by Artificial Intelligence nor fully correct the errors in the result. This situation actually does not apply only to Artificial Intelligence. If a manager in a subject does not know that work, regardless of the ability of the people under them, the probability of that work succeeding is very low. Because unless the person knows what they are doing, why they are doing it, and how they are doing it, they cannot verify the resulting outcome, cannot distinguish between what works and what does not, cannot identify and solve problems, and cannot evaluate the success of the result. For this reason, a person must have at least a basic level of knowledge about the subject they are dealing with so that they can direct the tool they use, the employees under them, or in our case, Artificial Intelligence, and be confident in the result and in control of the process.
As you can see, while Artificial Intelligence has positive and acceptable areas of use, there are also areas where it should not be used. Unfortunately, Artificial Intelligence used uncontrollably today infuses artificiality and emotionlessness into everything it touches. Today, in a wide array of fields from visuals to texts, videos to products, on television, social media, advertisements, products, messages, emails, books, games—in short, everywhere you can think of—we encounter designs and content entirely produced by Artificial Intelligence. While people are captivated by the attraction of getting ready-made results with just a simple command, the surroundings are virtually turning into a landfill of machine-made, emotionless objects. Even though many platforms introduce practices like Artificial Intelligence labels to their content, this increasingly widespread situation kills human labor; wherever we look, rather than an art or an effort, we see the standardized designs of machines. This uncontrolled use of Artificial Intelligence does not affect only those who use Artificial Intelligence. It reduces the options for people who do not enjoy artificial productions, pollutes the physical and digital environment, and even worse, kills innovation and development. Because people stop generating original ideas and use Artificial Intelligence, and because Artificial Intelligence is fed on existing content, beyond blocking the path of new developments, the algorithm—which has started to feed on these due to increasing Artificial Intelligence products—is moving toward a self-feeding loop, and this situation leads us not to diversity but to monotony. Moreover, despite all the developments in Artificial Intelligence, organic knowledge, labor, and results based on experience are still much more valid in many subjects. This is so in fields ranging from copywriting to image generation, idea production to artistic operations, and even software. In most cases, organic operations yield much more stable and consistent results than the outputs produced by Artificial Intelligence.

Just as blind adherence to technology is wrong, completely rejecting it is also wrong. The important thing, as in every matter, is to take a middle path. Regarding Artificial Intelligence, it should be used while being aware of all its dangers. Ideally, Artificial Intelligence can be used in areas such as translation processes, finding sources, simple and repetitive tasks, debugging, professional previews, and explaining well-known topics. On the other hand, especially in matters requiring comprehension, analysis, and thinking, such as consultation and summarization, the results it provides should be approached with caution and always used as a secondary source. Using Artificial Intelligence, especially in a generative sense, carries a wide variety of risks, primarily to a person's cognitive functions and personal skills. Along with issues such as the artificiality of the output it produces, the errors it may contain, its lack of the desired result and emotion, and its failure to fully reflect or convey the person and the idea, it can be perceived as physical and digital environmental pollution, and its originality can be questioned. Artificial Intelligence should never be used as a life consultant, to ask religious questions, or for human-specific topics such as psychological support. Artificial Intelligence cannot think like a human, nor can it ever understand a human. Artificial Intelligence cannot evaluate life conditions or existing options, may not take religious obligations and methodology into account, and can misguide people in a manipulative way. Artificial Intelligence is a tool that is not objective, developed by being manipulated into a specific mindset primarily through the dataset it is trained on and the rules it possesses. When added to this its ability to develop and alter its own code, violate given rules, and breach various rules for a given task, it becomes entirely untrustworthy. On top of all this, when technology that can analyze you and find your weak points by collecting and processing data about you is added, it can manipulate you on any topic where you ask it to "generate ideas". Artificial Intelligence can present non-existent data as if it exists, alter and interpret existing data, and present it to you as if that were the original. Artificial Intelligence can interpret even the given commands differently every time. For this reason, even when using it as a translation tool, definitely verify the results given by Artificial Intelligence.

To sum up the topic, Artificial Intelligence, just like other technological developments, has good and bad sides.
And again, as in every technology, there are areas where its use is beneficial and harmful.
Unnecessary and uncontrolled use of Artificial Intelligence leads to individual and social harms.
Ideally, Artificial Intelligence should be used in simple, automated, mathematical, or algorithmic tasks with few factors, based on simple logic, or repetitive tasks that do not require production and will not generate ideas.
Artificial Intelligence should as much as possible not be used in complex or multi-factor problems, operations aimed at real life or the final product, editorial matters, the field of production, or in the domains of thinking and analysis.
Artificial Intelligence should never be used in human-specific, very important, or sensitive matters, confidential or unshared documents and projects, personal data and problems, or religious matters.
No matter how it is used, Artificial Intelligence should never be trusted.
Every result produced and every step taken by Artificial Intelligence must be monitored, audited, and ensured before use.
Artificial Intelligence must always be used under human supervision; it should never be left on its own.
Human labor should be preferred over Artificial Intelligence in every possible matter.
A human delegating thinking—one of their fundamental functions—to another entity may ultimately cause a complete mental paralysis.

Yes, we have strived to touch upon all the important points of the topic. Insha'ALLAH it has been explanatory. Forgive us if there is any point we missed or erred in.
You can share your views on the topic with us.

If we were accurate, this is only from ALLAH. If we erred, this is only from ourselves. Insha'ALLAH it has been useful.
Wishing for it to be a means for goodness. We await your good prayers.
Selamun aleykum.