Introduction
In 2024 private U.S. investment in AI was $109.1 billion (Liberto) — an enormous sum that begs the question: will the returns from AI match these investments or are we in the middle of an unstable economic bubble? Hype around AI, fueled by various innovations, has caused these large AI investments. Many believe that AI is dangerously overvalued, fuelling fears of an AI Bubble, where the value of AI companies will plummet dramatically (Wheeler). However, others argue that these investments reflect AI’s potential, as something that can be transformative as the Industrial Revolution itself.
How We Got Here
Although early forms of AI, like chess machines, have existed since the 1900s (Mucci), AI only exploded into the mainstream in 2023 (Associated Press). This was largely due to OpenAI’s release of ChatGPT in late 2022 (Associated Press), which was a breakthrough in Generative AI—also known as GenAI—a form of AI that can create new content such as text and video (Stryker and Scapicchio).
ChatGPT displayed GenAI’s human-like capabilities, initiating rapid global adoption and investments because GenAI seemed like it had the potential to revolutionize the economy by skyrocketing people’s productivity and replacing workers (Goldman Sachs).
Reasons Supporting the Existence of an AI Bubble
However, Deepseek’s R1 large language model from December 2024 challenged the rationale behind these huge investments, the belief that major AI breakthroughs required massive investments in premium hardware and data centers. Deepseek matched or outperformed rivals like ChatGPT on benchmarks like problem-solving and math. This strong performance was despite it being built with less advanced Nvidia H800 chips and $6 million, just 5-10% of the cost of OpenAI’s $100M+ models (Turner). This caused tech stocks to decrease noticeably (Pappas) because investors interpreted this as evidence of an AI Bubble, where firms like Nvidia rose to trillion-dollar valuations based on hype around demand for expensive AI infrastructure.
Similarly, a report from MIT in July 2025 found that 95% of the 300 analyzed GenAI adoptions had zero returns, despite $30-40 billion in GenAI investments (Challapally et al. 3). Only 5% of the analyzed companies benefitting from GenAI adoptions despite such large investments suggests that AI investment won’t pay off. If this trend continues, the values of AI companies will likely plummet, which supports the idea that AI is overvalued.
Reasons Opposing the Existence of an AI Bubble
However, a study from September 2025 shows that on average, the 4,867 tested software developers using GenAI had a 26% increase in completed tasks (Cui et al. 1). And with increasingly powerful AI hardware like from Nvidia (Dion) and GenAI’s improving performance (Stanford Institute for Human-Centered Artificial Intelligence), there is potential for GenAI to become more and more useful in real-world applications.
According to Stanford economist Erik Brynjolfsson, new technologies follow a “productivity J-curve,” where initial deployment challenges lead to a disruption phase characterized by reduced productivity, followed by significant gains as integration improves. He uses electricity as an example, which became available in the 1880s but only boosted productivity after Henry Ford revolutionized factory production in the 1910s (Karma). Because of the speed AI evolves and its potential to increase productivity, we may merely be in the disruption phase of AI adoption, with exponential benefits coming soon.
Illustration of the productivity J-curve concept
(Flexible Production).
Conclusion
Whether we’re in the middle of an AI bubble or on the brink of a workplace revolution, one thing’s certain: AI is here to stay, and it’s changing how we work, think, and create. So instead of bracing for impact, maybe it’s time to lean in and ask: how can we shape this future to work for all of us?
Works Cited
Associated Press. “AI Went Mainstream in 2023. It Was Also the Year We Started to Panic.” Euronews, 27 Dec. 2023, www.euronews.com/next/2023/12/27/2023-was-the-year-ai-went-mainstream-it-was-also-the-year-we-started-to-panic-about-it. Accessed 26 Oct. 2025.
Challapally, Aditya, et al. The GenAI Divide: State of AI in Business 2025. Project NANDA, MIT, July 2025. MLQ.ai, mlq.ai/media/quarterly_decks/v0.1_State_of_AI_in_Business_2025_Report.pdf. Accessed 10 Oct. 2025.
Cui, Zheyuan (Kevin), et al. The Effects of Generative AI on High-Skilled Work: Evidence from Three Field Experiments with Software Developers. SSRN, 20 Aug. 2025, https://ssrn.com/abstract=4945566. Accessed 27 Oct. 2025.
Flexible Production. “The J-Curve concept and its application in your production company.” Flexible Production, 31 Jan. 2020, www.flexibleproduction.com/the-j-curve-concept-and-its-application-in-your-production-company/. Image. Accessed 26 Oct. 2025.
Goldman Sachs. “Generative AI Could Raise Global GDP by 7%.” Goldman Sachs, 5 Apr. 2023, www.goldmansachs.com/insights/articles/generative-ai-could-raise-global-gdp-by-7-percent. Accessed 26 Oct. 2025.
Harris, Dion. “NVIDIA Blackwell Raises Bar in New InferenceMAX Benchmarks, Delivering Unmatched Performance and Efficiency.” NVIDIA Blog, 9 Oct. 2025, blogs.nvidia.com/blog/blackwell-inferencemax-benchmark-results/. Accessed 28 Oct. 2025.
Karma, Rogé. “The Atlantic.” The Atlantic, theatlantic, 7 Sept. 2025, www.theatlantic.com/economy/archive/2025/09/ai-bubble-us-economy/684128/. Accessed 26 Oct. 2025.
Liberto, Daniel. “Which Countries Are Investing Most in AI?” Investopedia, 2025, www.investopedia.com/countries-investing-the-most-in-ai-11752340. Accessed 4 Nov. 2025.
Mucci, Tim. “History of Artificial Intelligence.” Ibm.com, IBM, 21 Oct. 2024, www.ibm.com/think/topics/history-of-artificial-intelligence. Accessed 26 Oct. 2025.
Pappas, Stephanie. “Why DeepSeek’s AI Model Just Became the Top-Rated App in the U.S.” Scientific American, 27 Jan. 2025, www.scientificamerican.com/article/why-deepseeks-ai-model-just-became-the-top-rated-app-in-the-u-s/. Accessed 15 Oct. 2025.
Stanford Institute for Human-Centered Artificial Intelligence. "Technical Performance." The 2025 AI Index Report, Stanford HAI, 2025, https://hai.stanford.edu/ai-index/2025-ai-index-report/technical-performance. Accessed 27 Oct. 2025.
Stryker, Cole, and Mark Scapicchio. “What Is Generative AI?” Ibm.com, 22 Mar. 2024, www.ibm.com/think/topics/generative-ai. Accessed 26 Oct. 2025.
Turner, Ben. “Chinese Researchers Just Built an Open-Source Rival to ChatGPT in 2 Months. Silicon Valley Is Freaked Out.” Livescience.com, Live Science, 24 Jan. 2025, www.livescience.com/technology/artificial-intelligence/china-releases-a-cheap-open-rival-to-chatgpt-thrilling-some-scientists-and-panicking-silicon-valley. Accessed 15 Oct. 2025.
Wheeler, Kitty. “What Is an AI Bubble & Its Risks to Enterprise AI Strategy?” Aimagazine.com, Bizclik Media Ltd, 17 Sept. 2025, aimagazine.com/news/what-is-an-ai-bubble-its-risks-to-enterprise-ai-strategy. Accessed 17 Oct. 2025.
