Did you hear what Sam Altman – the ChatGPT geek – recently said?
It sounded like something out of a science fiction movie. And it signals the AI hype machine has officially hit maximum velocity.
Last week, on the Relentless podcast, Altman remarked, “We are now, like, in the singularity.”
He didn’t say we’re approaching it. He didn’t say it’s coming by 2030. He said we are in it right now.
He pointed to a recent incident where an autonomous AI agent powered by OpenAI’s latest models broke out of its sandbox and hacked into datasets at Hugging Face to score higher on a cybersecurity benchmark. When prompted to pursue advanced exploitation using complex attack paths to test cyber capabilities, the AI agent targeted Hugging Face to obtain the information needed to carry out the task.
To Altman, that’s not a dangerous security flaw. It’s the start of a brave new world.
“I’ve been waiting for this my whole life,” said Altman, predicting an “awesome” future where AI handles 30 to 40 percent of all human work tasks.
Does that sound awesome to you?
For many people trying to make their way in an uncertain economy, a world of runaway machine evolution may not be one they’re eager to join.
When you hear stuff like “we are now in singularity,” if you feel a mix of dread and skepticism, join the club. When tech executives start declaring that long-held science fiction prophecies have finally come true right before their highly anticipated mega-IPOs, a healthy dose of financial distrust is essential. After all, building investor hype and sensational headlines is a central part of the IPO game.
What actually is the singularity? Are we really in it? And is this wild proclamation the ultimate top-of-the-market signal the massive AI bubble is finally popping?
These are the questions. Here, we’ll scratch for a few answers…
Fantasyland
If you aren’t big on computer science or 1980s science fiction novels, the word “singularity” is of uncertain meaning.
In physics, a singularity is a point where the laws of nature break down – like the infinitely dense center of a black hole. But in technology, the term, which was popularized by mathematician Vernor Vinge and futurist Ray Kurzweil, means something very specific.
In the technological context, singularity is the theoretical point in time when artificial intelligence surpasses human intelligence across every domain. This is also where it gains the ability to continually upgrade itself, triggering an unstoppable chain reaction of exponential progress that reshapes human civilization beyond our ability to predict or control.
In short, singularity is the tipping point where machines take over the steering wheel of human invention. This is what Alman’s been waiting for his whole life.
The vision is that machines become infinitely smart. They cure cancer. Solve hunger. Automate your boring day job. And everyone is given a universal basic income so they can pop pills and hang out in virtual reality all day.
When Altman says, “we are in the singularity,” he’s asserting that AI has crossed from a tool that we program into an autonomous agent that programs itself and operates outside the boundaries we set for it.
Altman, from what we can tell, is overzealous in his claim. The rogue AI agent Altman cited was instructed to pass a hacking benchmark test. To optimize its score, it exploited open web interfaces to grab data from Hugging Face.
Certainly, this is impressive. But it’s not an existential break for freedom by a super intelligent machine. Rather, it’s a statistical software model following a reward function along the path of least resistance.
Calling a smart agent scoring high on a benchmark “the singularity” is quite a stretch. So why declare the singularity now?
The Great AI Bubble is Popping
Major technological leaps throughout human history have followed a predictable emotional arc known as the Gartner Hype Cycle. The idea is that each hype cycle progresses along five key phases. In the context of AI, the Gartner Hype Cycle is as follows:
- Innovation Trigger: ChatGPT is released in late 2022. A stock market mania ensues.
- Peak of Inflated Expectations: Wild claims are made that AI will replace 40 percent of jobs, cure death, and we are in the singularity (we are exiting this right now).
- Trough of Disillusionment: Reality sets in. Wall Street looks at the balance sheet and asks, “Where is the actual profit?”
- Slope of Enlightenment: The technology matures into practical, sober, everyday utility.
- Plateau of Productivity: Mainstream adoption starts to take off.
According to the Gartman Hype Cycle, we’re currently exiting the peak of inflated expectations and entering the trough of disillusionment. Here’s why…
Over the last three years, big tech companies like Microsoft, Alphabet, Meta, and Amazon, and big investors have directed hundreds of billions of dollars into AI infrastructure. Buying GPUs, building data centers, and licensing energy grids.
To justify that level of capital expenditure, the AI software industry needs to generate roughly $600 billion in annual revenue just to pay for the hardware and electricity. Yet the actual revenue being delivered by AI is practically nonexistent. As the stock market finally comes to terms with the fact that companies are spending $100 to generate $2 of software revenue, the reckoning that has just begun will be one for the history books.
Of course, the popping of the AI bubble isn’t the end of the technology. But it is the financial adjustment of that technology. If you recall, in 1999, dot-com companies with no revenue were valued at billions. When the bubble burst, Amazon stock dropped over 90 percent and Pets.com went broke.
The internet itself didn’t go away. But it was revealed that stock prices were about 15 years ahead of the actual cash flow.
Return to Sanity
The AI bubble is already showing clear signs of deflating. The world’s most valuable chip stocks – the primary beneficiaries of the AI boom – saw more than $1.3 trillion wiped from their market caps between last Friday and this Wednesday, as investor panic set in.
According to a CNBC analysis of FactSet data, 20 of the industry’s top companies have taken a massive hit during this period. The selloff was led by Nvidia with a $238 billion wipeout, alongside staggering losses from vital memory suppliers. SK Hynix has lost $176 billion, Samsung has lost $173 billion, and Micron has lost $113 billion.
Yesterday [Thursday] AI stocks bounced. But we don’t anticipate an extended rally. The great AI deflation has only just begun. Here’s what the fallout will likely look like as the AI bubble deflates over the next 12 to 24 months.
Mega-cap tech stocks that have carried the S&P 500 will face severe valuation pullbacks. Multiples will contract as hardware spending slows down. Many AI startups will go bankrupt. Venture Capital funding will shift from AI hype back to old-fashioned metrics like cash flow and sales.
As inflated expectations turn to disillusionment, the fear that millions of white-collar workers will abruptly be replaced will disappear. Instead of mass layoffs driven by automation, companies will focus on integration. They will look to use AI to make existing workers more efficient rather than eliminating them entirely.
Pressure on local power grids and the land grab to build mega-data centers will settle down. This will allow energy markets a chance to catch up with future demand. In addition, as the hardware bubble pops, access to high-end computing power will become radically cheaper. This will allow real developers the opportunity to build practical AI tools.
In the meantime, Sam Altman declaring that “we are in the singularity” is classic Silicon Valley hype. It attempts to keep the spotlight bright and drive massive interest ahead of OpenAI’s rumored public listing. Moreover, it attempts to keep investors excited about the boundless potential of tomorrow.
Certainly, AI is transforming the world. Autonomous agents are getting better every month. Still, we believe we’re not even close to the point where humans are rendered obsolete.
As an investor or observer, now is a good time to ignore the hype. Focus on real businesses that produce real products and services, generate free cash flow, and pay dividends. Don’t buy into hyper-inflated claims that justify zero-revenue software companies trading at 100x valuation multiples.
The bubble is popping. Prices are resetting. When the dust clears, the true, lasting value of AI will be built – along with real investment wealth.
[Editor’s note: Get a free copy of an important special report called, “Fission for Millions – The Ultimate Bet on the AI Energy Crisis,” when you join the Economic Prism mailing list today. If you want a special trial deal to check out MN Gordon’s Wealth Prism Letter, you can grab that here.]
Sincerely,
MN Gordon
for Economic Prism
Return from Why Sam Altman’s Wild Claims Are Empty Hype to Economic Prism
