Alright mate, pull up a stool. We need to talk about this AI madness, and I’ve got opinions.

---

So I’m sitting at this bar on a cruise ship a couple weeks ago– proper gin joint, the kind where the bartender knows your drink before you do – and this tech bro slides up next to me. Designer jeans, startup t-shirt, the whole uniform. Guy starts telling me about his AI company that’s “disrupting everything” and how they just raised fifty million at a billion-dollar valuation.

“What’s your revenue?” I ask, genuinely curious.

“Well, we’re pre-revenue, but we’ve got incredible user engagement metrics and—”

I stop him right there. “Mate, that’s like saying you’ve got incredible swimming technique but you’ve never been in water.”

And that, right there, is the entire AI market in 2025. We’ve collectively lost our minds.

Let Me Break This Down Over a Rum Runner!

Look, I’ve been around the block enough times to recognize a bubble when I see one. And brother, this one’s got all the classics – the breathless excitement, the “this time is different” mantras, and venture capitalists throwing money around like they’re at a strip club on expense accounts.

OpenAI just got valued at $300 billion on $10 billion in revenue. That’s a 30x multiple. For context, Apple – you know, the company that basically owns the world – trades at 7-8x revenue. Even Tesla at peak Elon-madness never hit those numbers.

But here’s the kicker that’ll really cook your brain: OpenAI lost $5 billion last year and won’t be profitable until 2029. They’re burning through cash like that crazy uncle Jimmy burns through wives – spectacularly and repeatedly.

A $300 billion company that won’t make money for four years. Let that marinate for a second.

Déjà Vu All Over Again

This whole thing reminds me of the stories about the dot-com boom. Same energy, same crazy eyes, same complete abandonment of basic arithmetic. Back then, we had Pets.com spending millions on Super Bowl ads to sell dog food online. The Nasdaq hit 5,000 in March 2000 before it all went to hell.

The difference? Dot-com P/E ratios hit the high 40s, while today’s sitting around 32.5x. Still nuts, but not completely unhinged. Plus, unlike the ’90s where most companies were just burning VC money, many AI leaders actually have real revenue streams.

Microsoft isn’t some startup dreaming in their garage – they’re funding this circus with actual cash flow. NVIDIA makes real chips that people need, even if they did lose $600 billion in market value in one day when some Chinese kids showed up with better math.

But still, man. The vibes are eerily similar.

The Chinese Kids Who Crashed the Party

Speaking of DeepSeek – these absolute legends came out of nowhere and pushed training costs down to less than 1/10th of what American companies were paying. Overnight, they made every premium AI service look like they’d been selling bottled water at concert prices.

NVIDIA’s stock took a $600 billion haircut in one day. SIX HUNDRED BILLION. That’s more than most countries’ entire GDP, gone faster than my paycheck after a Vegas weekend.

You know what that tells me? This whole market was built on sand and unicorn tears.

The Math That Makes My Head Hurt

Here’s what’s really keeping me up at night, and why I think we’re all collectively insane: tech companies are planning to spend $320 billion on AI infrastructure in 2025. That’s not investment money – that’s “bet the farm on magic beans” money.

David Cahn from Sequoia says companies need to generate about $600 billion in revenue to justify current AI spending. They’re nowhere close. Not even in the same zip code.

And here’s the beautiful part – AI isn’t like traditional software. With normal software, you build it once and it costs almost nothing to serve each additional user. AI? Every single interaction costs real money. It’s like the difference between owning a printing press and hiring monks to hand-copy every book.

OpenAI’s costs are growing right alongside their revenue, sometimes faster. That’s not scaling – that’s just getting bigger while staying equally broke.

Why This Feels Different (But Also Exactly the Same)

Now don’t get me wrong – I’m not some grumpy old bastard yelling at progress. AI is genuinely mind-blowing. I’ve read about it helping doctors save lives, helping writers find their voice, solve problems we couldn’t even describe five years ago.

The technology is real. But so was the internet in 1999, and that didn’t stop people from paying $200 for Pets.com stock.

What gets me is how we’ve completely abandoned basic business sense. When someone tells me their company is worth billions but won’t make money until the next decade, I want to ask: “What exactly are we buying here? Potential? Dreams? Really expensive lottery tickets?”

The Beautiful Madness of It All

But here’s the thing – bubbles aren’t entirely evil. They’re just markets having a nervous breakdown while trying to process revolutionary change. Messy? Absolutely. Occasionally catastrophic? You bet. But they also fund the future.

The internet bubble gave us fiber optic cables, Amazon, and Google. All that “wasted” money built the infrastructure for everything we use today. Same thing’s happening now – even if half these AI companies go bust, we’ll still have incredible technology and infrastructure.

The trick is knowing which horses to back.

My Completely Unsolicited Advice

Look, if you’re putting real money on this table, here’s my take: stick with companies that have strong EBITDA margins and scalable models. NVIDIA, despite the recent drama, still makes actual products. Microsoft has cash flow that could fund small wars.

The startups burning billions on potential? That’s venture capital’s job, not yours. Let the professionals lose their shirts on moonshots while you invest in the picks and shovels.

During the dot-com crash, investors who diversified beyond tech actually came out ahead. Just saying.

The Reality Check That’s Coming

Here’s what I know for sure: hype cycles always end in reality checks. Always. The excitement inflates values way beyond what makes sense, and when gravity kicks in, only the strongest survive.

We’re not there yet, but we’re getting close. Some AI categories are trading at 70x revenue multiples. That’s not valuation – that’s wishful thinking with extra zeros.

The question isn’t whether some of these companies are overvalued (they obviously are), but whether you’ll be smart enough to ride the wave or wise enough to get out before it crashes.

The Bottom Line (And Why I’m Not Losing Sleep)

So are we in a bubble? Probably. Does it matter? Not really.

This is just how humans process change – badly, expensively, but eventually we figure it out. The AI revolution is real, but revolutions are messy affairs. People get hurt, fortunes get lost, empires fall.

But something beautiful emerges from the chaos. Always does.

My advice? Enjoy the show, but don’t bet the house on it. We’re living through history – the kind future business students will study while shaking their heads and wondering what we were all smoking.

And honestly? I’m here for it. This beautiful, expensive madness is exactly the kind of thing humans do when we’re trying to build tomorrow. We overshoot, we crash, we learn, we rebuild.

It’s like my grandfather used to say: “Son, the only difference between a boom and a bust is timing and who’s holding the bag when the music stops.”

So keep dancing, but know where the exits are.

Now, another round? Because watching artificial intelligence try to justify artificial valuations is thirsty work, and I’ve got more stories about Silicon Valley’s beautiful insanity…

---

What’s your take on all this? Think I’m right, wrong, or just having too much fun watching the circus? Drop a comment – I love a good argument almost as much as I love a good whiskey.