AI Bubble or the New Dotcom? How Businesses Can Make Money on the Looming Crisis | News Idea

Analysts are sounding the alarm: the AI market is repeating the patterns of the dotcom bubble. The S&P 500 index is growing thanks to a handful of giants, while $700 billion in data center investments does not guarantee profits. But it is precisely in instability that the most profitable strategies are born.

What happened

Financial analysts are increasingly warning of a bubble in artificial intelligence, comparing the current situation with the notorious dotcom boom of the late 1990s. A key indicator is alarming: at the record close of the S&P 500 in May, only 20 out of 500 companies reached their own all-time highs. The rest of the growth is being driven by the semiconductor sector and the so-called Magnificent Seven: Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla.

The combined capitalization of this seven exceeded $35 trillion. At the same time, companies are spending unprecedented sums on infrastructure: in 2026 alone, spending on AI data centers will reach $700 billion. Startups Anthropic and OpenAI are preparing IPOs, attracting billions of dollars. The historical parallel is frightening: in 2000, internet companies collapsed by 77%, leaving behind vacant lots in Silicon Valley.

How this is useful for business

A bubble is not a threat, but a window of opportunity for those who know how to read the market. During the dotcom crash, companies with real business models survived: Amazon, which then seemed unprofitable, became a trillion-dollar giant. Now it is a similar moment for the AI sector: when speculative capital leaves, technologies with proven utility will remain.

Current volatility creates arbitrage opportunities. While corporations overpay for GPUs and cloud services, small businesses can gain access to advanced tools at reduced prices. Investors who invested at the peak are looking for liquidity: this is the time for advantageous acquisitions and partnerships.

How to make money on this

The earning strategy is built on risk asymmetry. While major players accumulate positions in overheated assets, smart capital is preparing alternative scenarios. First, hedging through short positions on overvalued AI stocks. Second, investments in second-order infrastructure: energy suppliers, cooling systems, network equipment, which grow regardless of whose model wins.

The third path is creating businesses that save clients money on AI infrastructure. When the cost of computing decreases (and it will decrease after the correction), demand for optimization will soar. The fourth tactic is acquiring talent at reduced prices after inevitable layoffs in the tech sector.

Business ideas

1. AI consulting for SMB. Medium-sized businesses do not understand how to implement AI without insane costs. Consulting on solution selection, integration with existing systems, and staff training. Revenue: fixed rate $5,000–$15,000 per project plus a subscription fee of $1,000–$3,000/month.

2. Marketplace of prompts and templates. Creation and sale of ready-made solutions for ChatGPT, Claude, Midjourney for specific tasks: lawyers, doctors, real estate agents, HR. Model: one-time purchase $10–$50 or subscription $19/month. Margin 80%+.

3. AI expense audit. A consulting service for companies that overpay for cloud GPUs. Infrastructure optimization, transition to cheaper models, reduction of unused capacity. Commission 20–30% of the client's annual savings.

4. Training and certification. Courses on working with AI tools for corporate clients. Corporate contracts from $20,000 per group. Additional revenue: creating methodological materials that can be sold repeatedly.

5. AI task outsourcing platform. A marketplace where freelancers offer services for working with AI: text generation, data analysis, visualization. Commission 15–20% per transaction. Scales through automation of moderation and quality checks.

6. AI job monitoring service. An aggregator of hiring data in artificial intelligence with analytics on salary, skills, and geography trends. Subscription for recruiters $99–$299/month, advertising integrations with HR brands.

Risks and limitations

The main danger is incorrect timing. The bubble may continue inflating for years, and short positions will destroy capital before the market collapses. Regulatory pressure is also unpredictable: authorities may introduce restrictions on AI development, which would crash the entire sector at once.

The technological risk is real: current models may turn out to be a dead end, while quantum computing or new neural network architectures will change the rules of the game. Competition in consulting and training is intensifying: entry barriers are minimal, and margins attract thousands of new players.

7-day action plan

Day 1–2: Research the market. Identify 3–5 niches where AI solutions save clients time or money. Conduct 10 interviews with potential customers about pain points and willingness to pay.

Day 3–4: Choose a business model. Estimate starting investments, required competencies, and time to first revenue. Define a minimum viable product.

Day 5: Create a landing page or prototype. Launch lead collection to test demand before writing code or purchasing equipment.

Day 6: Start small. Launch the service for 3–5 clients for free or for a symbolic fee in exchange for a case study and recommendations.

Day 7: Calculate the economics. Determine customer acquisition cost, average check, and payback period. Adjust positioning and pricing.

Original news: Fast Company · See other news in the news section.