Are We In An AI Bubble?

Key Concepts

AI Bubble: The central theme, exploring whether the current rapid growth and investment in Artificial Intelligence is unsustainable and prone to a significant downturn.
Dot-com Bubble: Used as a historical comparison to the current AI landscape, highlighting similarities and differences.
Bull Case (Against AI Bubble): Arguments suggesting AI is not in a bubble due to strong financial backing, existing infrastructure, and current utility.
Bear Case (For AI Bubble): Arguments suggesting AI is in a bubble due to lack of profitability, unfulfilled revolutionary potential, and questionable financial practices.
Roundtripping: A financial practice where companies fund each other's growth in a circular manner, potentially inflating valuations.
Recession vs. Bubble Pop: Distinguishing between a bubble causing a recession and a recession causing a bubble to pop.
Generative AI: AI capable of creating new content, such as text, images, and videos.
Artificial General Intelligence (AGI): Hypothetical AI with human-like cognitive abilities.
GPU (Graphics Processing Unit): Specialized processors crucial for AI training and inference.
S&P 500: A stock market index representing 500 of the largest publicly traded companies in the United States.

The AI Bubble Debate: Arguments For and Against

This video presents a comprehensive analysis of the arguments surrounding whether the current surge in Artificial Intelligence investment constitutes an "AI bubble." It contrasts the current situation with the dot-com bubble of the late 1990s and explores the potential consequences if a bubble were to burst.

Bull Case: Why AI is NOT a Bubble (or Not Yet)

The "bull case" argues that the current AI boom is fundamentally different from past speculative bubbles and is built on solid foundations.

Strong Financial Backing:
- Unlike the dot-com era where money flowed into unproven startups with no clear path to profitability (e.g., pets.com), current AI investment is primarily from established, highly profitable tech giants like Google, Microsoft, Amazon, and Meta.
- These companies are reinvesting profits from existing successful businesses, not taking on debt. For example, Microsoft is not going into debt to fund AI development.
- Despite significant AI spending (hundreds of billions of dollars), these companies continue to post record profits, indicating financial resilience.
Tangible Utility and Usefulness:
- The internet in the late '90s was full of promise but lacked immediate widespread utility. AI, conversely, is already useful today.
- Millions of people use AI daily for tasks like writing, coding, image generation, and job automation.
- Even if not perfect, AI is demonstrably making tasks faster, easier, and saving time and money for individuals and companies.
Established Infrastructure:
- AI development is not starting from scratch. It leverages existing, robust infrastructure built over decades, including cloud computing, global data centers, powerful GPUs, and extensive software innovation.
- This stable foundation means that even if individual AI projects fail, the underlying technological infrastructure remains intact.
Resilient Funding Sources:
- The companies driving AI innovation are trillion-dollar giants with substantial capital, talent, and global infrastructure, capable of sustained investment.
- This contrasts with desperate startups solely reliant on venture capital for survival.
Real Value Being Built:
- A bubble pops when there's no underlying real value. The argument is that AI is creating tangible value, even if its full impact takes time to materialize. The hype, for once, might be directed towards something genuinely significant.

Bear Case: Why AI IS a Bubble

The "bear case" presents counterarguments, suggesting that the AI boom exhibits characteristics of a speculative bubble.

Lack of Widespread Productivity Gains:
- A significant study indicated that approximately 95% of companies integrating generative AI reported zero productivity improvement, with some experiencing a decline. Only about 5% saw a positive return.
- This suggests that many businesses are rushing into AI out of fear of missing out ("FOMO") rather than genuine, profitable application.
- Usefulness (generating text/code) does not automatically equate to profitability or problem-solving.
Gap Between Promise and Reality:
- AI was hyped as revolutionary, with talk of Artificial General Intelligence (AGI). However, current applications like AI girlfriends, image generators, and video tools (e.g., OpenAI's Sora 2, used as a "glorified meme generator") are seen as impressive but not world-changing.
- Even advancements like Google's AI-powered search are considered minor upgrades rather than transformative changes.
- The argument is that AI is not yet solving deep structural problems in science, medicine, or education.
Massive Unprofitability of AI Companies:
- Companies like OpenAI, Anthropic, and even Google's AI divisions are largely unprofitable.
- Running large language models (LLMs) is extremely expensive due to massive computing power and electricity consumption for every prompt.
- For instance, OpenAI reportedly loses money on every free ChatGPT prompt, and paid plans are insufficient to cover operational costs.
The Nvidia Dynamic and "Roundtripping":
- Nvidia is an exception, booming by selling the GPUs and chips that power AI, acting as the "shovels in the gold rush."
- A concerning financial dynamic has emerged:
  - Nvidia invests $100 billion in OpenAI.
  - OpenAI commits to spending a large portion of that on Nvidia chips.
  - OpenAI also promises AMD significant chip purchases and receives 10% of AMD's stock.
  - Oracle secures a $30 billion deal for cloud services for OpenAI.
- This creates a circular flow of money where companies fund each other's growth, making valuations appear inflated without genuine independent profit generation.
- This reliance on investment rather than profit is seen as unsustainable. If investments dry up, the entire ecosystem could collapse.

What Happens When the Bubble Pops?

The video explores two scenarios for a bubble burst:

Bubble Causes Recession: The bubble grows so large that its collapse drags down the entire economy (e.g., the 2008 housing crisis).
Recession Pops the Bubble: An economic slowdown makes investors nervous, leading them to withdraw cash from high-risk sectors like AI, thus popping the bubble.

Current Data Suggests Scenario 2:
- Approximately 30 AI-invested companies constitute 44% of the S&P 500's total value.
- Reports suggest the AI boom is masking underlying economic weakness. In 2025, around 40% of US real GDP growth is attributed to tech (AI-related) spending. Without it, the US economy would have grown by only 0.1% in the first half of 2025.
- Some analysts warn that stripping away AI spending might already indicate a recession.
Potential Consequences:
- If the economy slows and investors pull money from AI, major companies (Microsoft, Google, Nvidia, Amazon) are highly exposed.
- Smaller startups would likely shut down, leading to widespread layoffs.
- The industry could shrink dramatically, leaving only a few dominant players.
The Irony and Reset:
- The companies that created the AI bubble might be the only ones large enough to survive its pop.
- Bubbles can also serve as a reset mechanism, clearing out noise and revealing true value. If AI is a genuine revolution, it will likely survive, albeit in a different form.

AI's Own Opinion

When presented with the video's script and asked for a one-word answer to "Are we in an AI bubble?", ChatGPT responded after 12 seconds with: "Probably."

Conclusion and Call to Action

The video concludes by emphasizing that regardless of the AI bubble's fate, human intelligence, curiosity, and the willingness to learn remain irreplaceable. It promotes the Alux app as a tool for personal growth and wealth building, offering lessons, insights, and expert courses. The message is to invest in education, as AI's evolution will not negate the importance of personal learning and development.