# SIMPLEST Explanation of How Artificial Intelligence Works? No Jargon | What is AI? How AI works?

https://www.youtube.com/watch?v=m8o2GrbR3d8

[00:00] Should we be afraid of artificial
[00:01] intelligence? It is a question that
[00:04] keeps many of us up at night. Most
[00:06] people first heard about AI through
[00:08] 1990s movies like Terminator or The
[00:11] Matrix, where it was portrayed as a
[00:13] villain turning against humanity. And it
[00:15] was not just the movies. Renowned
[00:18] figures like Steven Hawking have also
[00:20] warned about the possible dangers of AI
[00:22] rising against human beings. So back
[00:25] then it was completely natural to feel
[00:28] uneasy whenever we heard the term
[00:30] artificial intelligence. But for nearly
[00:33] 20 years after those movies, AI made
[00:35] little real world progress that could
[00:37] cause serious concern. Things stayed
[00:39] relatively quiet. But today the
[00:42] situation has completely changed. AI is
[00:45] now part of our daily lives through AI
[00:48] cameras, Alexa, Siri, ChatGpt, Google
[00:52] Gemini, and many more tools. Even the
[00:54] YouTube algorithm that suggested this
[00:56] video to you works using AI. And it is
[01:00] not just YouTube. On platforms like
[01:02] Facebook and Instagram, AI decides what
[01:04] you see on your feed. Another unsettling
[01:07] use of AI is in deep fake videos, which
[01:10] many of you may have already come
[01:12] across. In several countries, banks now
[01:14] use AI to approve loans. Insurance
[01:17] companies use it to calculate your
[01:19] premium, and AI even plays a major role
[01:22] in stock market trading. In the medical
[01:25] field, AI is now being used for
[01:28] diagnosing diseases. In short, AI is
[01:31] being used in areas where serious
[01:33] decisions are being made. Yet despite
[01:36] all this, we have not seen AI behaving
[01:39] like the evil villain shown in those
[01:41] science fiction movies. And that
[01:43] naturally leads to an important
[01:45] question. Is there a difference between
[01:47] the AI we saw in movies and the AI we
[01:50] are seeing today? What exactly is
[01:52] artificial intelligence? How does it
[01:55] work? How many types of AI are there?
[01:57] And could AI ever become a villain as
[02:00] shown in the films? Let us explore the
[02:02] answers together in this video.
[02:08] Hi friends, welcome to a new video from
[02:11] science simplified for all. Most of us
[02:14] have been familiar with AI based
[02:15] personal assistants like Alexa, Siri and
[02:18] Google Assistant for some time now. But
[02:20] it was only after the arrival of Chat
[02:22] GPT and Google Gemini that many people
[02:24] truly realized just how far artificial
[02:27] intelligence has evolved. Chad GPT seems
[02:30] to understand what we say in natural
[02:32] language and respond in a way that feels
[02:34] remarkably human. Because of this, many
[02:37] people unknowingly assume that chat GPT
[02:40] is a conscious being, something with
[02:42] self-awareness. But we must always
[02:44] remember Chat GPT is just a software
[02:47] program specifically designed to mimic
[02:50] the way a human respond in conversation.
[02:52] It does not truly understand things the
[02:54] way a human does. To clearly grasp this
[02:57] idea, we first need a basic
[02:59] understanding of what artificial
[03:01] intelligence actually means and how it
[03:03] works. This video will give a very
[03:06] simple explanation aimed at everyday
[03:08] viewers with no technical background or
[03:11] prior knowledge about AI. So if you are
[03:14] someone already familiar with the core
[03:16] concepts of AI, feel free to skip this
[03:19] part as it may seem like an
[03:21] oversimplification.
[03:23] Let us begin with the term intelligence.
[03:26] The ability to learn new things, make
[03:29] logical decisions and solve problems.
[03:32] That is what we call intelligence in a
[03:34] human being. When a machine or computer
[03:36] begins to exhibit this kind of ability,
[03:39] we call it artificial intelligence. That
[03:42] is the most basic definition of AI. But
[03:46] this definition alone does not give us
[03:48] the full picture. We already know that
[03:50] computers have been capable of doing
[03:52] mathematical calculations much faster
[03:55] than humans for decades. For example, if
[03:58] you ask a computer to multiply 1,230
[04:01] by 2480,
[04:03] it will do it instantly, far faster than
[04:06] any human ever could. In fact, a
[04:09] computer can perform millions of such
[04:11] calculations every second. In that
[04:13] sense, computers have always been ahead
[04:16] of us in solving arithmetic problems.
[04:18] But here is the key point. These
[04:20] calculations are all based on
[04:22] instructions that we the humans have
[04:25] already given to the computer in
[04:26] advance. These instructions are what we
[04:29] call programs. A standard computer can
[04:32] only follow those predefined
[04:34] instructions exactly as given. If it
[04:36] encounters a situation slightly
[04:38] different from what we programmed it
[04:40] for, it will usually fail. Now, here is
[04:43] the difference. When a computer is able
[04:45] to do something new, something we did
[04:48] not specifically teach it by learning
[04:50] from the data and recognizing patterns
[04:52] on its own, that is when we say the
[04:55] computer has artificial intelligence.
[04:57] Let us take an example. Suppose we
[04:59] create a program to identify birds and
[05:02] upload it into a computer. Then we show
[05:04] the computer pictures of 10 different
[05:06] types of birds. A crow, an eagle, a
[05:10] sparrow, a parrot, and so on. We also
[05:13] tell the computer these are all birds.
[05:16] Now imagine we show the computer a new
[05:19] picture of a bird it has never seen
[05:21] before. If the computer is able to
[05:23] recognize that the new image is still a
[05:26] bird even though we never showed it that
[05:28] specific example, that is a simple form
[05:31] of artificial intelligence. It has
[05:33] learned from the previous examples and
[05:35] applied that knowledge to something new.
[05:38] That is one of the key traits of AI. Of
[05:41] course, what today's artificial
[05:43] intelligence is capable of goes far
[05:45] beyond this simple example. AI systems
[05:47] now handle much more complex and
[05:50] powerful tasks. But we are using this
[05:52] bird example here to make the core
[05:54] concept easier to understand. Now, let
[05:57] us take a closer look at how the
[05:59] artificial intelligence program in our
[06:01] bird example actually works. We begin by
[06:04] defining the unique features that
[06:06] distinguish a bird from other objects.
[06:09] For example, we might say that birds
[06:11] usually have two legs, wings, feathers,
[06:14] and a beak. These are the features that
[06:16] we define in the program. Next, we
[06:19] assign a weightage to each of these
[06:21] features. That means we decide how
[06:23] important each feature is in identifying
[06:25] a bird. For instance, two legs 20%,
[06:29] wings 30%, feathers 30%, beak 20%.
[06:35] Together that adds up to 100%. Now if we
[06:39] show the computer a picture of a crow
[06:41] which has all four features it would
[06:44] score 100% and the program would
[06:46] confidently say yes this is a bird. But
[06:50] suppose we show it a picture of a
[06:52] penguin. A penguin does not have visible
[06:54] wings or feathers like other birds. Its
[06:57] wings look more like arms. So the
[06:59] program might only find 40% of the
[07:02] features it is looking for. Based on
[07:04] this low score, it may wrongly decide
[07:08] this is not a bird. But we know that a
[07:10] penguin is a bird. So we correct the
[07:13] program and tell it that it made a
[07:15] mistake. When the program receives this
[07:17] correction, it begins to adjust the
[07:19] weights it had assigned earlier. It
[07:21] might reduce the importance of features
[07:23] like feathers and wings and increase the
[07:25] weightage for features like a beacon two
[07:28] legs. It also learns that a perfect 100%
[07:31] match is not always necessary. Even if a
[07:34] creature matches only 80% of the key
[07:37] features, it could still be a bird. This
[07:39] program is written in such a way that it
[07:42] can adjust these weightages by itself
[07:44] based on feedback. As we continue to
[07:47] show the program more and more images of
[07:49] different birds, it keeps refining its
[07:51] internal settings. Each time it makes a
[07:54] mistake, we correct it and the program
[07:56] learns from that error. In other words,
[07:59] it adjusts its internal parameters.
[08:02] Eventually, after seeing enough examples
[08:04] and making enough adjustments, the
[08:06] program becomes capable of correctly
[08:09] identifying any bird, even one it has
[08:12] never seen before. This process is what
[08:14] we mean when we say an AI program is
[08:17] being trained. In reality, the working
[08:19] of an AI system is far more complex than
[08:22] what we just explained. Modern AI
[08:25] programs use many different parameters
[08:27] and weight values and these are
[08:29] processed through multiple layers and
[08:31] stages before reaching a final decision.
[08:34] Still, we chose to present the concept
[08:36] in such a simple way for one important
[08:39] reason to help you understand three key
[08:42] points clearly. First, in the example we
[08:45] discussed, the AI does not actually
[08:48] understand what a bird is. In fact, the
[08:51] concept of understanding or mind itself
[08:53] is quite vague. But even if we leave
[08:55] that aside, what a human understands
[08:58] when they see a bird is fundamentally
[09:00] different from what an AI sees. For the
[09:02] AI, a bird is just a group of numbers,
[09:05] weight values associated with different
[09:07] features like a beak, feathers, and
[09:10] other such traits. That is all. Second,
[09:13] to train an AI, we need a massive amount
[09:16] of data. Even in the bird example, the
[09:19] AI needs to be shown thousands of bird
[09:21] images to learn what a bird looks like.
[09:24] And not just that, every time the AI
[09:27] makes a mistake, a human has to step in
[09:29] and correct it, telling the AI that it
[09:32] was wrong. Without that feedback,
[09:34] learning will not happen. Third, even
[09:38] after all this effort, the AI only
[09:40] learns to answer one specific question.
[09:43] Is this a bird or not? If you want the
[09:46] AI to answer a different question like
[09:48] which bird it is, you need to write a
[09:51] different program and train it
[09:52] separately. And if the image is not a
[09:55] bird at all and you want the AI to say
[09:57] whether it is a mammal or something
[09:59] else, that too requires a separate
[10:01] program and new training. In other
[10:03] words, each AI can only perform the
[10:06] specific task it was trained for. If you
[10:08] want the AI to do a new job, it must be
[10:11] trained again, often from scratch. To
[10:14] put it simply, every specific task
[10:17] requires a specially trained AI system.
[10:20] Just because something is called
[10:21] artificial intelligence does not mean it
[10:23] can do everything. All these limitations
[10:26] we discussed so far explain why even
[10:29] though science fiction movies talked
[10:31] about AI decades ago, the real
[10:33] development of artificial intelligence
[10:35] took much longer to happen. The rapid
[10:38] growth of AI that we see today became
[10:40] possible mainly because of two key
[10:42] factors. The first reason is the
[10:45] dramatic increase in computer processing
[10:46] power. Back in the 1990s, computers
[10:50] simply did not have the capability to
[10:52] run AI programs. But over the last 20
[10:56] years, the computational speed of
[10:58] computers has increased by several
[11:00] thousand times. This incredible
[11:03] improvement is what finally made it
[11:05] possible to run complex programs like
[11:08] artificial intelligence on regular
[11:09] machines. The second major factor was
[11:12] the rise of social media. Let me give
[11:14] you an example. When you upload a photo
[11:17] of your pet dog to Facebook and tag it
[11:19] as dog, you're actually helping
[11:21] Facebook's artificial intelligence learn
[11:23] what a dog looks like. Every time
[11:26] someone does this, the AI gets better at
[11:29] recognizing dogs, even when it sees a
[11:31] totally new photo of a dog later. With
[11:34] the explosion of social media, AI
[11:36] systems suddenly had access to huge
[11:38] amounts of data for training. Today,
[11:41] millions of people across the world
[11:43] upload photos every single day. Many of
[11:46] those photos are used behind the scenes
[11:48] to train AI programs. The same is true
[11:51] for public messages, comments, and
[11:53] captions that we post. All of this
[11:56] content is used to help train artificial
[11:58] intelligence in understanding natural
[12:01] language. the way we humans actually
[12:03] speak. And this is important because the
[12:06] way we speak is very different from what
[12:08] you find in books or dictionaries.
[12:10] Spoken language is filled with informal
[12:13] phrases, slang, and regional styles. Yet
[12:16] today, we have reached a point where AI
[12:18] can understand meaning even in informal
[12:21] speech. And that is largely thanks to
[12:24] the vast amount of training data AI has
[12:27] received from language used on social
[12:29] media. By now you should have a general
[12:32] idea of how an artificial intelligence
[12:34] system is trained. But there are a few
[12:36] important things you need to know. When
[12:39] an AI is trained using millions of
[12:41] images and messages, the process is not
[12:44] done manually by humans. That would be
[12:46] practically impossible. Instead, the
[12:49] training is handled automatically by
[12:52] powerful software systems designed for
[12:54] that purpose. And this leads to a major
[12:56] problem. By the time training is
[12:58] complete and the AI is released for
[13:00] public use, even the developers who
[13:03] created the system often have very
[13:05] little idea of what exactly is happening
[13:08] inside the AI's decision-making process.
[13:11] This is because the training process
[13:13] changes the AI program so extensively
[13:16] that it becomes difficult to track how
[13:18] it arrives at its conclusions. Here lies
[13:21] the real issue. If such an AI makes a
[13:24] mistake, it is incredibly difficult to
[13:26] figure out where it went wrong or why it
[13:29] made that mistake. This lack of clarity,
[13:32] often called the blackbox problem, is
[13:34] one of the biggest drawbacks of many
[13:36] modern AI systems. Let me give you a
[13:39] real world example. An AI system was
[13:42] trained to identify animals in pictures.
[13:45] But during testing, it started
[13:47] mclassifying certain dog photos as
[13:49] wolves. At first, the reason for this
[13:52] behavior was unclear. Only after
[13:54] detailed investigation did researchers
[13:56] discover the actual cause. In the
[13:59] training data set, almost all wolf
[14:01] images had snow in the background. So,
[14:04] whenever the AI saw snow behind a dog,
[14:06] it assumed the image must be of a wolf.
[14:09] In other words, it was giving more
[14:11] importance or weight to the snowy
[14:14] background than to the animal itself.
[14:16] This kind of issue happens because after
[14:19] training we no longer fully understand
[14:21] what the AI is focusing on internally.
[14:24] That is why it becomes so difficult to
[14:26] trace and correct these kinds of errors.
[14:29] To address this problem, a new approach
[14:31] called transparent AI has been proposed.
[14:35] The idea is to make AI systems more
[14:38] understandable where we can see and
[14:40] interpret what is happening inside.
[14:42] However, transparent AI is still a
[14:45] developing concept and has not yet been
[14:47] widely adopted in real world
[14:49] applications. There is one more crucial
[14:51] point to remember. The data used to
[14:54] train an AI must be completely accurate
[14:57] and unbiased. If the training data
[14:59] contains flaws or biases, those issues
[15:02] will reflect directly in the AI's
[15:04] behavior. Let me give you a real world
[15:06] example. In one case, a company used an
[15:09] AI system to shortlist candidates for
[15:11] job interviews by analyzing job
[15:14] applications. But when the results came
[15:16] out, it was found that the AI had
[15:18] selected only male candidates. On
[15:21] investigation, the reason became clear.
[15:24] The data used to train the AI mostly
[15:26] came from past hiring decisions. And in
[15:29] the past, the company had preferred
[15:31] hiring men for that specific job role.
[15:34] This bias in the historical data, even
[15:36] if unintentional, got passed on to the
[15:39] AI system. As a result, the AI learned
[15:43] to favor male candidates simply because
[15:45] that is what the past data showed. This
[15:48] is a classic example of how bias in
[15:50] training data can lead to discrimination
[15:52] in AI decisions. And this can apply to
[15:55] other human biases as well, such as bias
[15:58] based on skin color, religion, or any
[16:01] number of other prejudices. That is why
[16:03] it is absolutely essential that the data
[16:06] used to train AI is carefully checked
[16:09] for fairness and neutrality. Otherwise,
[16:11] our own biases will get transferred into
[16:14] the AI we build. And here is something
[16:16] we must never forget while using AI.
[16:19] When an AI makes a mistake, it has no
[16:22] idea that it made a mistake, nor does it
[16:24] feel any regret. That is because AI has
[16:28] no concept of understanding. It does not
[16:31] know what is right or wrong. It simply
[16:33] follows the patterns it was trained on.
[16:35] So the responsibility to monitor AI
[16:38] behavior will always rest on us the
[16:40] humans. We must be the ones to watch to
[16:43] correct and to decide what is
[16:45] acceptable. Now let us look at the
[16:47] different types of artificial
[16:49] intelligence used today and the kinds of
[16:51] jobs they are designed to do. The first
[16:54] major type is called natural language
[16:56] processing or NLP. This refers to AI
[17:00] systems that can understand and respond
[17:02] in human language, the way we actually
[17:04] speak. Examples you're already familiar
[17:07] with include Alexa, Siri, and Google
[17:10] Assistant. But NLP is used in many other
[17:13] areas as well beyond just virtual
[17:15] assistants. The second type is
[17:18] generative AI. This is a kind of AI that
[17:21] can create new content, things that
[17:23] never existed before. For example,
[17:26] imagine an AI that has read thousands of
[17:29] novels. Now, if you ask it to write a
[17:31] completely new novel, it can do that. Or
[17:35] if it has seen thousands of human faces,
[17:37] you can ask it to generate a picture of
[17:39] a face that does not belong to any real
[17:42] person and it will do that too. That is
[17:45] what generative AI does. Chat GPT and
[17:48] Google Gemini are examples of generative
[17:51] AI. To be more specific, they are
[17:54] generative text AI, which means they
[17:57] create new text based on what they have
[17:59] learned from existing text data.
[18:01] Similarly, there are generative image
[18:03] AI, which can create new images from
[18:06] learned visual data. One such tool is
[18:09] Deli. For instance, you could ask it if
[18:12] a famous painter from the past were
[18:14] alive today, what would their painting
[18:16] of a modern city look like? and the AI
[18:19] would generate an image based on that
[18:21] prompt. The third type is called
[18:23] computer vision AI. This kind of AI is
[18:27] used for image and face recognition. AI
[18:30] cameras which can identify people,
[18:32] objects or license plates fall under
[18:34] this category. Beyond these, there are
[18:37] many other specialized types of AI
[18:39] available today. Robotic AI helps robots
[18:42] navigate and interact with the world.
[18:44] Speech recognizing AI converts spoken
[18:47] words into text. Explainable AI is
[18:50] designed to make AI decisions more
[18:52] transparent. Planning and scheduling AI
[18:55] are used in logistics, project
[18:57] management, and more. In some cases, two
[19:00] or more types of AI work together to
[19:03] perform a more complex task. You might
[19:05] have seen deep fake videos where a
[19:08] famous actor's face is seamlessly
[19:10] swapped onto another person's body or a
[19:12] politician appears to say something they
[19:14] never actually said. This is possible
[19:17] because two AI systems are working
[19:20] together in tandem. The first AI system
[19:23] is trained specifically to swap faces in
[19:26] a video. But if you look closely at
[19:28] those videos, you might notice something
[19:30] odd, some slight unnaturalenness that
[19:33] makes you realize it is fake. Here comes
[19:36] the role of the second AI program. Its
[19:39] job is to detect flaws in the video.
[19:41] Anything that seems artificial or off.
[19:44] Once the flaws are identified, the first
[19:46] AI makes corrections and produces a new,
[19:50] improved version of the video. Then the
[19:52] second AI checks it again to see if any
[19:55] new mistakes are still visible. This
[19:57] process continues in multiple rounds
[19:59] with both AIs competing. One trying to
[20:02] make the video more realistic, the other
[20:04] trying to detect any imperfections.
[20:07] Eventually, the outcome is a deep fake
[20:09] video so convincing that we can no
[20:11] longer tell it is fake. The deep fake
[20:13] videos we have seen so far are not
[20:15] perfect. But today there are AI tools
[20:18] capable of generating even more
[20:20] realistic and convincing videos. This
[20:23] raises a disturbing truth. Such highly
[20:25] accurate deep fakes can potentially be
[20:27] used for fraud or other malicious
[20:30] purposes. So far we have been
[20:32] classifying AI based on the kind of work
[20:34] it does. Image recognition, text
[20:37] generation, speech processing and so on.
[20:40] But there is another way to classify
[20:42] artificial intelligence based not on the
[20:44] task it performs but on its level of
[20:47] capability. This is where we come across
[20:49] the terms weak AI, strong AI and super
[20:53] AI. And it is in this classification
[20:55] that the possibility of AI turning
[20:58] against humanity like in the movies
[21:01] enters the picture. The AI systems we
[21:03] have discussed so far are all designed
[21:06] to do one specific task. This kind of AI
[21:09] is known as weak AI or narrow AI. Almost
[21:13] all the AI that exists today falls into
[21:16] this category. We humans on the other
[21:19] hand possess what is called general
[21:22] intelligence. We can learn a wide
[21:24] variety of things, adapt to different
[21:27] situations, make logical decisions, and
[21:30] solve many types of problems. If a
[21:32] computer ever develops that kind of
[21:34] broad capability, it would be called
[21:36] general AI or artificial general
[21:39] intelligence. Sometimes simply called
[21:42] strong AI. But let us be clear,
[21:44] artificial general intelligence does not
[21:47] exist yet. It is still a theoretical
[21:50] concept. There are ongoing efforts to
[21:52] build such an AI, but no one knows how
[21:55] long it will actually take. Some experts
[21:57] in the field who are very optimistic
[21:59] believe it might take around 20 years.
[22:02] Others say it could take at least 50
[22:04] years or even more. As of today, general
[22:07] AI remains a future possibility, not a
[22:10] present reality. There is one more
[22:12] category of AI that is even more
[22:14] powerful than general AI. It is called
[22:18] artificial super intelligence. This
[22:20] refers to a future AI that will be more
[22:22] intelligent than humans in every
[22:24] possible way. Right now, this kind of AI
[22:28] is purely speculative, but that does not
[22:30] mean it can never happen. Some believe
[22:33] that such an artificial intelligence
[22:35] might emerge by the end of this century,
[22:38] but no one really knows how long it will
[22:40] take. The main concern is this. Once
[22:44] artificial super intelligence is
[22:45] created, it could potentially become
[22:47] smart enough to design even better
[22:50] versions of itself. And those versions
[22:52] could go on to create even more advanced
[22:55] versions and so on. If that happens, the
[22:58] growth of AI would become exponential
[23:02] like a chain reaction. This kind of
[23:04] situation is what experts refer to as
[23:06] the technological singularity. It is
[23:09] this level of super intelligent AI that
[23:11] is often portrayed in movies as turning
[23:13] against humanity. And it is not just in
[23:16] fiction. Many respected individuals have
[23:18] expressed concerns about this
[23:20] possibility. The late physicist Steven
[23:22] Hawking and Elon Musk, CEO of Tesla and
[23:25] Space X, have both openly warned about
[23:28] the potential dangers of uncontrolled
[23:30] AI. But there are also many experts who
[23:33] believe such fears are unfounded, at
[23:35] least for now. Their main argument is
[23:38] this. Today's AI systems do not have
[23:41] consciousness or self-awareness. As we
[23:43] discussed earlier, when an AI identifies
[23:46] a bird, it does not truly understand
[23:48] what a bird is. It is simply processing
[23:52] numbers and patterns. It has no inner
[23:54] awareness or sense of meaning. In fact,
[23:57] even human consciousness is still a
[23:58] mystery. We do not fully understand what
[24:01] consciousness is or how it arises in the
[24:03] brain. So, creating a conscious
[24:06] self-aware AI is far beyond our current
[24:09] capabilities. And without
[24:10] self-awareness, the idea of an AI
[24:13] deciding to take control or destroy
[24:15] humanity is not realistic. After all,
[24:18] such desires for power and domination
[24:21] are very much human traits. Even if an
[24:23] AI becomes smarter than humans, it does
[24:26] not necessarily mean it will want to
[24:28] harm us. What we should actually be
[24:30] concerned about is not the AI itself,
[24:33] but the humans who might misuse it. AI
[24:36] is a very powerful tool and like any
[24:39] powerful tool it can be used for good or
[24:41] for harm depending on who is using it.
[24:44] That is the real risk we must watch out
[24:46] for. In fact, some of this may already
[24:49] be happening. AI can be used to subtly
[24:51] influence the general public toward a
[24:54] particular political orientation or
[24:56] ideology. And with many of the AI
[24:58] algorithms we interact with in daily
[25:00] life, achieving this kind of influence
[25:02] is relatively easy. Another common fear
[25:05] people have about AI is that it will
[25:09] take away jobs. But we have faced this
[25:12] situation many times in history. When
[25:14] electricity and machines were
[25:16] introduced, people feared job losses.
[25:19] When computers came, there were similar
[25:21] fears. Yet each time humanity adapted.
[25:25] We found new ways to work and new types
[25:28] of jobs were created. Many experts
[25:30] believe the same will happen with AI.
[25:33] Artificial intelligence will not
[25:34] directly take away your job. But someone
[25:37] who learns to use artificial
[25:38] intelligence effectively might out
[25:40] compete you in the same field. That is
[25:43] where we need to adapt to the changing
[25:45] world. We must learn how to work with
[25:47] AI, not against it. I hope this video
[25:50] helped you gain a clear and simple
[25:51] understanding of what artificial
[25:53] intelligence really is and what it is
[25:55] not. If you found this video useful,
[25:57] give it a like and share it with someone
[25:59] who might find it interesting, too. And
[26:02] if you enjoy content that explains
[26:04] complex science in a way that anyone can
[26:06] understand, make sure to subscribe to
[26:08] the channel and of course tap the bell
[26:11] icon so you do not miss any of our
[26:13] upcoming videos. We have some
[26:15] fascinating topics coming soon. Thank
[26:18] you.