Ah, the AI market. If we’re talking about the digital world’s latest obsession, it’s like watching a really fancy car zoom by driven by a blind man – and just like with blockchain and NFTs, we all know there’s going to be a crash. But for now, everyone’s convinced that this shiny new toy is the future. And sure, maybe it is, but there’s a whole lot of smoke and mirrors mixed in with the promises of world domination. So let’s buckle up.

 

The Rise: AI as the Next Big Thing (Again)

Let’s not kid ourselves – AI is the latest shiny object that everyone’s getting their hands on, and I’m not just talking about your AI-generated cat pictures. No, I’m talking about the flood of companies, venture capitalists, and tech giants all scrambling to claim their slice of the AI pie. Remember when AI was just a few buzzwords from nerdy engineers and the tech-curious? Now, it’s in everything – from your emails to your recommended Netflix shows, and apparently, it’s also going to rewrite how we do business.

 

AI, of sorts, has been around for decades, but 2023 was its coming-out party. Suddenly, we were all talking about large language models (LLMs) like GPT, deep learning, and machine learning, as if the world had just discovered fire. And to be fair, these tools are pretty impressive—when they work. But much like blockchain and NFTs, the hype around AI has been cranked up to 11.

 

The market grew exponentially. In 2024 alone, the global AI industry was estimated to be worth around $300 billion, and everyone was waving it around like the hottest stock on Wall Street. AI startups were getting funded by the boatload, and everyone was cashing in on the potential. AI-powered this, AI-generated that—heck, even AI-designed companies were getting off the ground. Just like blockchain promised to decentralize everything, AI promised to automate everything.

 

The Fall: AI Overload – Did We Overhype It Again?

Okay, so we’ve seen this script before, right? A shiny new tech enters the scene with a promise of world domination, and within two years, it’s all crumbling down with a chorus of “Oh, maybe we got a little carried away.” The AI market, my friend, is in danger of going down that same road. You’ve heard of the classic startup crash-and-burn, right? AI is getting dangerously close to that, except this time, the stakes feel higher because now everyone’s talking about how AI will change everything from healthcare to education, and even creativity.

 

Yes, AI is impressive—but it’s not magic. Sure, AI can write, translate, analyze, and even pour a pint (well, maybe not that last part… yet). But here's the thing: it still has some serious growing pains. There’s still a massive gap between the potential of AI and its actual capabilities. In fact, a lot of the tech we are currently gushing over is more “sizzle” than sausage. For every fantastic deep learning model, there are a thousand more that are borderline useless, often needing massive amounts of data to perform anything close to competent.

 

The promises are heavy. In some ways, it’s like 2017's blockchain hype all over again. AI will "revolutionize" industries, but at the end of the day, it’s still mostly the same AI models running on complex algorithms that only serve to confuse the heck out of most people trying to use them. And as for its actual business applications? Well, don’t get me started. Sure, automation sounds sexy in theory, but in practice, it’s a big job security scare for the masses and a race for efficiency that’s only going to leave a lot of people feeling replaceable.  But then I suppose it also opens up roles for AI maintenance, training, operation etc… Evolution.

 

The Middle Ground: AI Sticks Around, But Not for the Reasons You Think

So, what’s the deal with AI moving forward? I think we’re going to see it stabilize into something more sustainable—but that doesn’t mean it’s all roses. AI is here to stay, but it won’t be the overblown, sky-high predictions we’re hearing now. It’s probably going to follow the blockchain/NFT trend – but with a twist.

 

Here’s where AI is actually useful: content generation, data analytics, process automation, and even in specialized fields like healthcare and finance. These applications? They’ve got legs. But unless we’re talking about specific use cases, a lot of the “next big thing” chatter is just that—chatter.

 

And this annoys me - if robots are generating content (Chat GPT), robots are reading and summarising content (Kaizan), does this make the human-human interaction at either end more or less efficient?

 

The market will likely start to look like a few established players (hello, OpenAI, Google, Microsoft), who dominate the high-value AI applications. Then there’ll be some niche AI startups riding the coattails of these big players, attempting to be the next big thing in AI-generated art, video editing, or chatbots. But just like in every industry, only a few of them will make it, and the rest will fold under the weight of their overblown valuations and lack of real-world applications.

 

It’s all going to come down to the execution of the tech. If companies can pull off legitimate AI-powered tools that solve real problems, AI will be as ubiquitous as smartphones, probably without anyone even thinking about it. But if we’re stuck with fluff like NFT profile pictures and chatbots that sound like they belong in a sci-fi novel, well, let’s just say we’re going to be having that conversation in 2025, too.

 

The er, Media Ground: Just don’t tell me how it works..

It’s like we’re all just spinning in this tech-saturated circle where the focus has been on how things work instead of why we should care. And let’s be real, in the world of media buying, this has been a big issue, especially with programmatic. If you remember the early days, media buyers were bombarded with “the tech,” “the algorithms,” and all these complex terms about “machine learning” and “bid density.” Cool, but that’s not why we’re buying media, is it? It’s because we need results, not a crash course in how artificial intelligence or a bidding system works.

 

Programmatic, like blockchain and NFTs, fell into the trap of focusing too much on the “how.” We got so caught up in explaining the underlying tech—how data is pinging across exchanges and how bids are placed milliseconds before an ad shows up—that we forgot about the most basic question: what is it actually doing for the media buyer, and moreover, the client? And let’s be honest, for most people in media buying, the answer was murky at best. It wasn’t about doing anything tangible for them; it was about the promise of more efficiency and targeting. But efficiency and targeting alone don’t keep clients happy.  They’re looking for impact, conversions, and return on investment. If the tech isn’t solving that problem, then why should they care how it works?

 

Now, let’s talk blockchain. For a while, every conversation I had about blockchain felt like a tech conference rather than a media discussion. “It’s decentralized! It’s transparent! It’s secure!” All true, sure, but how does that actually impact media buys? What does that mean for a brand’s campaign performance? How does it make life easier for an advertiser? Instead of answering those questions, blockchain just kept doubling down on how cool the tech is, without considering if it was even solving an existing problem in media. And NFTs? Well, they were the ultimate example of tech without a solid answer to what it’s actually for.

 

And now we’ve got AI, which I fear is just more of the same. Sure, it’s doing things for media (like automating content creation, improving targeting, and predicting trends), but rather than talking about the value it adds to a campaign, everyone’s busy discussing how an AI model is trained, or how it learns, or how much data it needs to function. Should we care about how it works?

We care if it can help optimize ad placements, boost performance, and streamline workflows.

 

It’s like we’ve all been swept up in the magic of the tech without stopping to ask, Does it actually make my job easier or my client’s budget work harder?

 

We’re still trying to get media buyers to buy into AI, but if we don’t focus on the practical, tangible benefits—like how AI can reduce wasteful spend, boost engagement, and improve targeting—then we risk ending up in the same place we were with programmatic, blockchain, and NFTs: with all the promises and none of the payoff.

 

It’s time to stop talking about the how and start focusing on the what. Because in the world of media, if it doesn’t do something real and measurable, then what’s the point?  Go and have a look for youself at AI generated motivational posters - it needs finesse.

 

 

The Future: AI – More Than Just a Buzzword, but Not a Savior

So, where’s AI heading? The market may never see the same explosive growth that blockchain or NFTs had. It’s already starting to cool down, especially as the reality of AI’s limits is setting in. But AI’s future isn’t about everyone riding the hype wave and thinking it’ll replace human intelligence (spoiler alert: it won’t). Instead, it’ll likely become an embedded part of our daily lives—the quiet assistant that makes everything run just a little more smoothly without us having to think too much about it.  And that assistant who can help us figure out the burning questions we all have like ‘who is the most stereotypically metal band of all time?’*

 

Ultimately, AI’s future won’t be about what it can do, but how well it can do it. It’ll either find a useful, sustainable role in society, or it’ll end up being a glorified back-end service like cloud storage—important, sure, but not something we obsess over.

 

But for now? AI’s got that same overinflated ego we saw with blockchain and NFTs, and that, my friends, is a recipe for an inevitable crash. The question is: how long until we see it?

 

And I really don’t think we’re going to be under the threat of Skynet.

 

*it is W.A.S.P.