Mid-week Decaf

The AI Bubble, Investor's FOMO and the selling of prospects instead of value

In partnership with

Ever since ChatGPT was launched there has been a frenzy in the tech world. The much anticipated future of AI was finally here. Or was it? Anyway, the idea of a future powered by artificial intelligence caused an upheaval in the tech world. Investors threw their support behind the development of AI leading to what many experts refer to as a Bubble.

A bubble is an economic phenomenon where an asset is valued at a price that is far greater than the intrinsic value of the asset

Looking at the development of artificial intelligence in the past two years, it is undeniable that some progress has been made. But it hasn’t been in leaps which is understandable since the technology is still relatively new. However, investors seem to want to accelerate the pace of AI development beyond the capability of the technology at this time, throwing millions of dollars at AI any new AI startup. This decision seems to be motivated by fear of missing out (FOMO), and who would blame them? AI has minted new billionaires, the likes of Sam Altman and Jensen Huang, while previous billionaires who made an early bet on the technology also gained massively.

The side-effect of the AI gold rush

AI Startups are not strapped for cash. This will normally be a good thing, but excess of everything can be bad. One of the side-effects of the AI gold rush is that the excess cash seems to be dampening its development. Although we have seen many big tech firms battle for the top spot in the AI race, there still hasn’t been any meaningful progress post-ChatGpt. Each iteration of AI meant to compete against ChatGpt has been a flop. Beginning with Google’s Gemini AI to the latest, Devin, the AI software engineer. Despite the glamorous pre-launch videos demonstrating the AI in action, users have been left in utter disappointment after realizing that the AI is not close to performing at the level seen in the videos. The videos were staged!!!

We are talking about millions of dollars in investments and the only result is a half-baked demo. Clearly, the idea here is simple to sell the prospects of what can be done, instead of real value. Is it possible for AI to achieve the level of complexity shown in these videos? Maybe. The question now is when would that be? Two years? Five? or perhaps 10? How long are investors willing to keep funding these companies? To put things into perspective, here is how much it takes to train an AI.

Get value stock insights free.

PayPal, Disney, and Nike recently dropped 50-80%. Are they undervalued? Can they recover? Read Value Investor Daily to find out. We read hundreds of value stock ideas daily and send you the best.

When OpenAI released GPT-3 in 2020, cloud provider Lambda suggested the model—which had 175 million parameters—cost over $4.6 million to train. OpenAI hasn’t disclosed the size of GPT-4, which it released a year ago, but reports range from 1 trillion to 1.8 trillion parameters, and CEO Sam Altman vaguely pegged the training cost at “more than” $100 million. Anthropic CEO Dario Amodei suggested in August that models costing over $1 billion would appear this year and “by 2025, we may have a $10 billion model.

Source: Yahoo Finance

The trend doesn’t seem to end with AI software like Gemini and Devin only. AI products like the Humane AI pin and the Rabbit R1 both failed to deliver anything remotely close to the hype that trailed the launch of the products. Tech YouTuber Marques Brownlee summed it all up when he called the Humane AI Pin the “Worst Product” he has ever reviewed. He equally talked about the Rabbit R1 calling it “Barely Reviewable”.

So why is the influx of cash dampening the development of artificial intelligence?

Lack of competition. One of the factors that drives development is competition. Startups are traditionally very competitive because they need to prove themselves worthy of the money they are asking for. Why would an investor choose to invest in Startup A over Startup B? Well, it will typically boil down to how much value they see in each company. This means the two startups vying for funds will compete to deliver as much value as possible if they are to stand a chance of being funded. AI startups do not face this sort of pressure since they can easily have access to funding when needed. This can take away the innovative edge that comes with a traditional startup ecosystem.

Opinions: “AGI will happen this year” - AI researcher David Shapiro

So presently, AI can perform some tasks very well but still lag behind humans in many other tasks. However, AGI or Artificial General Intelligence should be able to perform several tasks at the human level. This is why it is commonly called Strong AI. Nvidia CEO Jensen Huang was previously quoted saying AGI will be achieved in 5 years.

"I would say that within the next five years, you're gonna see, obviously, AIs that can achieve those tests,"

Jensen Huang, Nvidia CEO

But Shapiro believes it will be coming this year with the launch of GPT 5. Here are some features of ChatGPT that are anticipated;

  • Significant improvement in language translation and humanlike interactions

  • Reduced AI hallucinations

  • More customizations

  • More processing power, &

  • multilingual support

Read more about GPT 5 features below;

So what is your take on this? Do you think we will finally see AGI in action with the launch of ChatGPT later this year?