How Intelligent is Artificial Intelligence? Debunking the AI Hype

Remember that time when AI-generated art won a competition, beating all the human-made art? Or when an AI-generated Fake Drake Song went viral? These moments indicate that Artificial Intelligence (AI) has not only evolved enough to match human creativity but, in some cases, take over and even outperform it, so much so that for the first time in human history, people can say with conviction that AI is really intelligent! 

This “intelligence” has not come without worries. With the recent flood of AI-generated content (images, reels, and videos) taking over social media, humans are having a tough time distinguishing what is real and what is AI-generated. This made us all realize another grave fact: that AI can not only take over and outperform humans, but also become a threat to us. 

But wait, before we get carried away with the fearmongering about AI controlling us, there’s a deeper question we need to ask: How Intelligent is Artificial Intelligence? The question might sound weird, but it tries to ask how something artificial can be truly intelligent. If AI is a tool to serve human convenience, how can it be so intelligent? And if so, how intelligent must it be to qualify as an artificial intelligence? 

This article will answer these particular questions. I will share my perspective on whether AI is really capable of sentient and intelligent thoughts or is just a sophisticated tool to recognize patterns and analyze data. And if the latter is true, maybe AI has not really evolved, but the hype was carefully crafted by the big tech companies to increase power and profit. 

What Big Tech Is Actually Selling?

AI was already present in some basic form in our lives, from small automotive tasks that were embedded in popular software or tools to adaptive algorithms that personalized our social media feeds. But it was just that, mere features in tools claiming to ease complicated tasks and provide convenience. It wasn’t AI, but it served the same purpose that AI claims to serve. 

Our idea of an Artificial Intelligence actually acting as an intelligent creation developed when ChatGPT first rolled out in November 2022. This realization that an AI has been developed that is good enough not only to assist and ease our tasks but also to potentially take over the entire process came when people started using ChatGPT to write entire essays and articles with just a simple prompt. This was the first time that we could ask an artificially created product/service something, and it would reply intelligently like a human. Since then, we have had many versions of ChatGPT and other AI tools that replicated human creativity with near perfection, making it almost impossible to separate them from human works. 

Rebranding AI Tools As “Thinking Machines”

This shift wasn’t just technological, it was also rhetorical. Big tech companies didn’t just release better software and better updates, they released a narrative. One that rebranded these AI models as being “sentient”, “autonomous”, “reasoning”, and even “empathetic”. Just look at how these AI models were introduced by the big tech companies in their keynote addresses:

• OpenAI - “Superhuman reasoning abilities” 

• DeepMind - “General intelligence” and “self-improving agents”

• Meta’s FAIR team on LLaMA - “AI that thinks like us”

• Other AI startup pitches - “Emotionally aware AI” or “empathetic responses”

These companies hijacked the AI narrative to promote their models as being one of a kind that can think and act like a human. In reality, none of these systems actually “feel”, “understand”, and “reason” in the way humans do. They’re simply good at recognizing patterns and guessing the next word, pixel, or frame based on the prior data. But the illusion of intelligence works, and that’s what the big tech companies are selling; not just tools of convenience but the dream of intelligence itself. 

Anthropomorphizing for Profit

Big Tech companies anthropomorphize AI tools deliberately by giving them human-like qualities, not because they have them, but because they help sell products/services, and more importantly, vision. These companies don’t need their AI models to be actually intelligent, they just need to give the impression of possessing intelligence, enough to satisfy customers and investors to ride the hype. 

If a product can write poems, songs, movie scripts, talk like a friend, give emotional support, or stimulate conversations, it gives the illusion that you’re conversing with something that has a mind. This illusion is profitable. It creates a supportive ecosystem for big tech companies where:

• Investors throw money at startups that can sell a futuristic AI vision.

• Consumers stay glued to these AI models because they feel “relatable” and “supportive”.

• Brands take advantage of this hype by integrating with AI and promoting themselves as “innovative”.

Basically, what these big tech companies are selling is not intelligence, but an illusion of intelligence. They just need to package these AI models as humane because it eliminates disbelief in their vision.

So What Are These Companies Actually Selling?

If big tech is just selling the illusion of intelligence, let’s look at what these companies promise and what they actually deliver:

Case 1: OpenAI’s ChatGPT

Marketing: “Superintelligence is coming.” OpenAI has promoted ChatGPT 4 as capable of understanding and reasoning about complex ideas. 

Reality: A large pattern-matching system trained on public data, fine-tuned to mimic human speech, that doesn’t understand, but it predicts. 

What’s being sold: Developer access, enterprise APIs, premium subscriptions, all powered by the myth of machine creativity.

Case 1: Google’s Gemini

Marketing: “Toward Artificial General Intelligence”. AI models like Gemini are described as approaching “human-level intelligence”.

Reality: Basically, an in-house feature that predicts your next steps and assists in completing those tasks. It’s also limited by its data and statistical weights.  

What’s being sold: AI as a medical diagnostic tool, as a code generator, as a scientific partner, all behind the guise of “general intelligence.”

Essentially, the big tech companies are selling machines that don’t think but act like they do. Through a carefully curated marketing campaign, they’ve managed to control the narrative that their AI tools or models are capable of human-level intelligence, when in fact, it’s a large pattern recognition machine that feeds from a vast database. In the following sections, we will look at how these AI models are able to maintain the veil of intelligence through human speech. 

What AI Really Is? Peeking Behind the Curtain

So, if AI is not really intelligent, then what is it? What’s with all the hype? What exactly is beneath all that branding and buzzwords? Let’s pull back the curtain.

The straight answer is that AI is a statistical prediction machine. In simple terms, it doesn’t think, understand, or feel what its users are asking. It just recognizes patterns in your queries and regurgitates back answers that are based on its training on a massive amount of data, like texts, images, and videos. It analyzes your queries and matches them with its database to predict the most appropriate response. 

For example, when you ask something on ChatGPT in English, it doesn’t actually understand English. It doesn’t even understand your query. What it does is to decipher each letter in your prompt and compare it with a multitude of similar patterns from its training data, and then accurately predict the next most probable word. This process is repeated until it forms a response that seems intelligent, sprinkled with some human emotions to make it more appealing. 

Artificial Intelligence As Mere Stochastic Parrots?

Notable linguist and AI critic, Emily Bender, described Large Language Models (LLMs) like ChatGPT as stochastic parrots. This means these models “probabilistically” link words and sentences together without comprehending their meaning. This lack of comprehension is the key thing that proves the unintelligent nature of Artificial Intelligence. Because just like a parrot, these systems mimic the style, tone, and language, without actually understanding the meaning and essence of the content they generate. 

This is terrifying because AI systems' lack of understanding of their own creation indicates that they also lack the ability to verify their own content. Moreover, AI systems are unable to hold themselves accountable for the process by which they arrive at an answer. 

Another leading voice in AI criticism, Gray Marcus, claimed that  “AI still lacks 'common' sense, 70 years later”. In his article, he iterated that today’s AI lacks common sense, understanding, and flexible reasoning, the essential things that define intelligence. 

All these point to the harsh truth that AI doesn’t understand the world, nor is it capable of intelligence. It merely finds patterns.

Red Flags of AI That Make It Anything But Intelligent

The AI hype is real. But like many other hypes out there, this too is without much substance, because other than some flash frontend and a few tech gimmicks (like Ghiblification), there’s an entire list of (intelligent) things that AI cannot do: 

• Lack of contextual understanding: AI can still not understand the entire context of a query. It doesn’t get sarcasm or cultural references, unless they are specifically trained on millions of examples of those styles.

• Lack of self-awareness: AI systems are not conscious or sentient. They cannot think for themselves and can’t comprehend the concept of ‘I’ or ‘self’. They don’t have any intentions, desires, or even the slightest idea of what they’re saying. 

• Unable to separate truth from falsehood: AI will say that the sun revolves around the sun if its database says so. It cannot think for itself, nor can it come to self-realization to understand right or wrong. This makes it dangerous as it will parrot lies or misinformation that has been fed into its database, without thinking for itself. 

• Vulnerable to biases and misinformation: Since AI systems can’t think for themselves, they rely on their trained database. Any biases or human error in that database will also affect the reliability of the AI systems and their ability to produce valid and accurate content. Basically, AI is as good as its database. 

These are just some of the important things that AI is unable to do now, or in the foreseeable future. However, these are the very things that make any artificial product/service intelligent. The essential qualities of contextual understanding, consciousness, emotion, and moral compass form the basis of intelligence. Lacking in any one of these qualities is a sure sign of unintelligence. Therefore, despite all the hype, Artificial Intelligence is, in fact, not Intelligent. 

So, if AI is not intelligent, what’s with all the hype? Why are we so excited about AI? Surely, many of us have realized that AI is not what big tech is selling, so why are we riding the bandwagon and calling for an AI revolution? We will explore these questions in the next section.

Why Did We Start Hyping AI In The First Place?

Despite the unintelligent nature of AI, both consumers and investors have not stopped being part of the hype. We are actively creating and participating in a global trend that venerates AI as the ultimate solution to every problem. We have raised AI to a pedestal of God-like intelligence, and started worshipping its different versions while standing in awe of its (super) human capabilities. Qualities like superfast, extreme efficiency, and precise prediction create an impression that AI models are intelligent, if not more intelligent, than humans. This impression is strong enough to stop us from questioning the intelligence of AI systems. So let’s delve deeper to deconstruct this unquestioning acceptance and explore the consumer psychology driving the AI hype. 

For this deconstruction, I will examine and track the AI hype from two perspectives: the consumer side and the Business side. Since these two segments have different interests and expectations from AI, it is better to understand each group’s perspective separately. So let’s begin.

Group 1: Consumer Side

We can’t blame ourselves for hyping AI, despite all the proof of it being anything but intelligent. Take this scenario: a freelance graphic designer is tasked with designing a couple of creatives for a marketing agency within 48 hours. An AI can do that within a few minutes, in a better and more efficient way. It’s the superhuman speed, the efficiency, and the accuracy with which it completes tasks that make people think AI is really capable, and thus the hype.

The truth is, for decades, we, consumers, have yearned for a new form of technology that resembles what we’ve seen in sci-fi movies or shows. AI is not a new concept; the main idea was introduced in the 1950s. However, the fact that we still hadn’t seen anything close to that concept materialize was a sign of slowing technological progress. This all changed when ChatGPT rolled out, opening a Pandora's box of exciting AI versions. First, it was only text-based models, then came image generators, and eventually, AI tools that virtually create all forms of content. It all happened in such a short time frame that we started getting AI models that exceeded our expectations.  Before we even have the time to process and experiment with a new AI model, another version rolled out. Soon, we got headlines on groundbreaking innovation in AI from around the world. We started getting new AI products/tools/services almost every month, some with similar features and some with exciting features. 

This sudden rush of AI models flooding the market, which almost eliminated mundane and repetitive tasks, caught the attention of billions of people. For the first time, we could ask a machine in a human language to perform a task, and it would do that faster and better than an actual person. Now, you might say that we had lots of new technological innovations before that eliminated mundane tasks and made our work a lot easier, and you’re right. Tools and products like Photoshop, Microsoft Excel, PowerPoint, and many more, eliminated so many mundane jobs that we don’t even remember. But what separates the AI innovation and the older ones is the way we use them. All the previous tools or technologies required us to actually do the work. It made processes easier and convenient, but we still had to put in effort to remove a stranger’s hand from the background of a photo in Photoshop, create corporate databases using complex formulas on Excel, and design informative and engaging slides on PowerPoint. The point is that we had to use those tools to produce an output. What’s different about AI is that it can produce the same output faster and more efficiently with a single prompt. We don’t actually use AI to create results; we just instruct it using plain language, and it does the work for us. 

On a psychological level, AI taps into our craving for shortcuts, ease, and control. The opportunity to get high-quality output with a single prompt that would’ve otherwise taken hours of hard work is not just convenient; it’s irresistibly rewarding. It appeals to our cognitive biases:

• Desire for instant gratification - Our brain's tendency to favor short-term rewards over long-term success.

• Automation bias - Our tendency to trust machine-made decisions over humans.

• Status Signaling - Using AI makes us feel we’re tech-savvy, forward-looking, and part of a global trend. 

• Fear Of Missing Out (FOMO) - Our fear of missing out on the AI trend, triggered by viral breakthroughs of constant AI innovation.

Add all these, and it’s no surprise that we, as consumers, have become so emotionally invested in the AI hype. Moreover, the constant improvement of AI models is reaching near perfection, and we understand why consumers are calling it an AI revolution. 

Group 2: Business Side

The business side behind the AI hype may seem different from the consumer side, but it’s the same story of desperation and yearning for something groundbreaking and era-defining. Similar to consumers, who desperately wanted some exciting and new piece of technology that would satisfy their curiosity, big tech companies also wanted to jump on the AI bandwagon. But the motivations were power, money, and dominance; something that would justify their stature and cement their status as global tech innovators.

A commonsensical viewpoint would be that since the invention of the iPhone back in 2007, technological innovation has stalled. But this would be an erroneous statement. We’ve had many tech innovations since the iPhone, but nothing groundbreaking in the consumer technology segment. This is a problem because all the top big tech companies have their main revenue stream from the consumer technology segment. Without any new and exciting innovation or invention in this front, they’ll eventually lose customers, leading to the stock prices going down. And if there’s anything that the business leaders and investors of these companies hate, is their company’s stock prices crashing. They will take a step to resolve this challenge, and they have taken it.

Cut to 2025, we are seeing new and exciting AI products/services almost every month. Companies and their investors are splurging a disgusting amount of money on AI projects, which they think would solve humanity’s biggest crisis (at least they’re marketing like it). Whatever the goal, they’ve cracked the code. Big tech knows that consumers have wanted technology like this for so long, and they will always be excited and hyped about it, as long as they get exciting new features (or believe that they’re getting these features). 

However, these big tech companies know that they’ve already plateaued in the AI front. What needed to be innovated or invented has already been done. All that is left is to perfect the AI models with new upgrades and features, but the core function of every AI system is the same. We are seeing this in real life, where ChatGPT has evolved from a text-based AI model to generating images and eventually, videos. It was a linear progression that was expected. ChatGPT didn’t evolve from a content-producing model to one that has the ability to control and manage an entire company by itself. That would be a real and exciting innovation, but we are centuries away from that right now. 

The point is, just like the iPhone and the onset of other smartphones more than two decades ago, there are only mere upgrades over the previous version. Every year, it’s a better and more polished version. All we see are products and services being perfected on, and that’s not a bad thing, but to hide behind those lazy upgrades and calling it an innovation is. These big tech companies are doing the same thing with AI. Instead of investing in an actual Artificial Intelligence system that at least resembles intelligence, we are seeing lazy and useless upgrades that serve no purpose. We are made to think that these AI models are not only intelligent but also ready to solve global problems. 

Enters the marketing department. Rather than spending huge sums of money on AI innovations, they spend an absurd amount of money on their marketing departments, with the specific purpose of creating and maintaining a hype around their specific AI model. The job of these marketing departments is to tap into the consumer’s psyche and feed them nearly useless AI features, packed in a shiny, minimalist frontend package, every month, while making them believe that these AI models will solve human crises. 

These marketing teams have realized that there’s an ample supply of similar AI models in the consumer market. This increases competition and decreases demand or hype over any single AI model. To combat this, they’ve aggressively run marketing campaigns that make their AI models, which aren’t even intelligent, look like they will end global hunger, bring global peace, and bring prosperity to all. Through slick presentations and short-form content advertising, they’ve created an image of an AI revolution that will change the world. This has become such a cliche that even most consumers are seeing through this charade. They overpromise, underdeliver, apologize, and repeat the cycle. 

This created an environment where short-term monetary gains take precedence over real innovation. Instead of focusing on new innovative AI models, startups are releasing half-baked products and slapping AI on them to secure funding. Investors see AI as the next gold rush, similar to Cryptocurrency, Bitcoin, and Web3, and pour money into anything that mentions AI. Similarly, big tech companies and their well-funded army of marketing teams completely rebranded their image to revolve around AI and its everlasting presence in our future by exaggerating AI’s capabilities to make products look essential. 

What True Artificial Intelligence Would Look Like?

If the big tech companies, who are supposed to lead the AI revolution, are not interested in creating true artificial intelligence, startups have given up. It begs the question of what a true artificial intelligence would look like. What is actually lacking in the AI we already have?

Simply put, true artificial intelligence will go beyond mere pattern recognition and data analysis to actually understand emotions, meaning, and context, and apply this understanding flexibly across unfamiliar contexts. A truly artificial intelligence creation would have the characteristics of common sense, consciousness, self-awareness, intuition, and the ability to think beyond the data it already possesses. It will be intelligent in a way that it can think for itself without any external support. This might sound like a borderline on science fiction, but this is what it takes to create true artificial intelligence, and anything less than that is not worthy of being called AI.

Today’s artificial intelligence is obviously nowhere near the illustration. What we have is essentially about statistical pattern-matching, which lacks real comprehension, understanding, curiosity, or self-driven goals. Despite groundbreaking achievements and billions of dollars of investment, what we call “AI” is just a product of clever marketing. AI today pretends to think without actually thinking. And until we bridge that gap from pretension to reality, AI will remain a powerful tool, but not truly intelligent. 

Final Thoughts: What Is The Future of AI?

We’ve established that Artificial Intelligence is not intelligent, at least not in the way we define human intelligence. We have also uncovered that the AI hype is less about reality and more about desperation, speculation, and corporate greed. But that doesn’t necessarily eliminate the “I” from the AI. On the contrary, AI is already embedded in our daily lives, and its influence will only grow from here. 

So the big question remains: Will AI replace humans? 

The short answer, yes and no.

As discussed in the article, the current AI systems are nowhere near capable of replacing a human job. Despite the grand claims, a truly intelligent AI is decades, if not a century, away. That said, some menial and repetitive tasks and roles have already been automated, and AI will continue that trend. But we have to understand that this is normal for any groundbreaking tool. Yes, AI is not really intelligent, but it’s still a powerful and capable tool that has the ability to optimize workflows and streamline entire work processes, saving both time and money. Therefore, AI will only swallow roles that are menial and repetitive, like small-scale data analysis, data entry, inventory management, etc. 

AI is still completely incapable of taking over job roles like content writing, designing, content creation, and more. These are the roles that AI can do with near perfection, but the keyword here is “near”. Doing these tasks requires human creativity, which AI lacks. Sure, it can optimize the job by taking up some of the mundane aspects of the job, but AI still can’t produce content that makes people cry and fall in love. 

The future of AI will be an AI-human collaboration, where AI will be like a smart assistant tool that will help us do our jobs by ourselves. It will give us clear directions, based on a human-created agenda, help us when we get stuck, and make our overall life a bit easier and convenient. But if you’re suddenly expecting AI to act like J.A.R.V.I.S., think again. We’re not there yet, and probably won’t be for a long time.