The first wave of AI was about possibility. The second wave is about profit.
Weiss Ratings just dropped a report claiming we are entering a new phase of AI market activity. They are calling it AI is Second Wind — citing 303% historical averages and fresh data signals that point beyond the first cycle.
Take it with the appropriate grain of salt. Weiss Ratings sells newsletters. But beneath the marketing hype, there is a real shift happening. The narrative around artificial intelligence is moving from look what this can do to show me the money.
And that changes everything.
The First Wave: Building the Cathedral
Go back to 2022-2023. ChatGPT launches. GPT-4 follows. The world discovers that AI can write essays, code software, generate images, and hold conversations that feel eerily human.
The market response was predictable: pure hype. NVIDIA stock went parabolic. Every company added AI to its pitch deck. Venture capitalists threw billions at anything with generative in the name. The 5 billion tech deal spree was not about immediate returns — it was about not missing out.
This was the infrastructure phase. Build the data centers. Train the foundation models. Establish the platforms. The assumption was simple: if we build it, the use cases will come.
And they did. But not fast enough to justify the valuations.
The Hangover: When Hype Meets Reality
By late 2024, the cracks were showing. AI spending was massive. Revenue? Not so much.
Enterprises were experimenting, not deploying. Pilot projects stayed in pilot. The tools were impressive, but integrating them into real workflows proved harder than the demos suggested. Microsoft is 3 billion bet on building its own AI stack started looking less like strategic genius and more like expensive necessity.
The market cooled. AI stocks corrected. The narrative shifted from AI will change everything to okay, but when?
This is where Weiss Ratings sees the opportunity. Their Second Wind thesis argues that we are not at the end of the AI boom — we are at the transition point between phases.
The Second Wave: Monetization Mode
Here is what the second phase looks like in practice:
Revenue over reach. Companies stop optimizing for user growth and start optimizing for revenue per user. OpenAI launches ChatGPT Enterprise. Anthropic focuses on Claude for business. The free tiers get stingier.
Vertical solutions over general platforms. Instead of AI that does everything, we get AI that does specific things extremely well. Legal document review. Medical diagnosis support. Financial fraud detection. Code generation for specific languages and frameworks.
Inference at the edge. The first wave was about training massive models in data centers. The second wave is about running smaller, specialized models on devices. NVIDIA is Jetson Thor is not just for robots — it is for every device that needs to run AI without calling the cloud.
Cost discipline. The spray and pray approach to AI R&D ends. Companies focus on use cases with clear ROI. The infrastructure buildout continues, but with more scrutiny on utilization rates and payback periods.
What the 303% Figure Actually Means
Weiss Ratings cites a 303% historical average for second-wave technology plays. Here is the logic: when a transformative technology emerges, the first wave captures the infrastructure builders. The second wave captures the application layer.
Think about the internet. The first wave was Cisco, Dell, and the telecoms building the pipes. The second wave was Google, Amazon, and Facebook building on top of them.
Or mobile. First wave: Apple, Qualcomm, the carriers. Second wave: Uber, Instagram, WhatsApp.
The infrastructure players get the early hype. The application players get the sustained returns.
Weiss is betting that AI follows the same pattern. The NVIDIAs and Microsofts of the world built the foundation. Now the companies that figure out how to actually make money with AI — the vertical applications, the workflow integrations, the enterprise solutions — are the next opportunity.
The Sectors to Watch
If the monetization thesis is correct, these are the areas where the second wave is already building:
Enterprise automation. Not replacing workers, but making them dramatically more productive. Document processing. Data entry. Customer service triage. The boring stuff that companies actually pay for.
Healthcare. Diagnostic assistance. Drug discovery. Administrative automation. The UK is regulatory push around AI in healthcare is not just about safety — it is about creating frameworks for deployment at scale.
Financial services. Fraud detection. Risk modeling. Algorithmic trading. Customer service. Banks have the data, the budgets, and the regulatory tolerance for AI that works.
Manufacturing and robotics. NVIDIA is physical AI push is not happening in a vacuum. Industrial automation is where AI meets the physical world, and the economics are compelling.
Crypto and DeFi. The infrastructure plays are following a similar pattern. First the platforms, then the applications. AI-powered trading, risk management, and compliance are the next frontier.
The Counter-Argument: Why This Time Might Be Different
Before you mortgage the house to buy AI stocks, consider the bear case.
Commoditization risk. Large language models are becoming commodities. Open source models (Llama, Mistral) are approaching closed-model performance. If everyone has access to similar capabilities, competitive advantage disappears and margins compress.
Regulatory headwinds. The EU AI Act. US state-level regulations. The UK is CMA investigation into Microsoft. Regulators are waking up to AI is implications, and compliance costs could eat into profits.
Economic sensitivity. The second wave thesis assumes companies keep spending on AI optimization even if the economy softens. That is not guaranteed. If we get a real recession, productivity enhancement budgets get cut fast.
The hype cycle. 303% historical returns sound great. But past performance does not guarantee future results. The sample size for transformative technology waves is small. The pattern might not hold.
What This Means for Investors
If you buy the second wave thesis, the playbook is clear:
Avoid the infrastructure plays. NVIDIA might keep growing, but the easy money is made. The risk/reward at current valuations is questionable.
Look for vertical applications. Companies solving specific problems for specific industries. The narrower the focus, the stronger the competitive moat.
Watch for profitability. In the first wave, revenue growth was enough. In the second wave, unit economics matter. Look for companies with clear paths to sustainable margins.
Consider the picks and shovels. If you are not sure which applications will win, invest in the infrastructure that all of them need. The AI infrastructure arms race is not over — it is just entering a different phase.
What This Means for Everyone Else
Even if you are not investing, the second wave matters.
Job disruption accelerates. The first wave threatened writers and artists. The second wave targets accountants, paralegals, customer service reps, and middle managers. The boring jobs that pay the bills.
Productivity gains become real. The first wave was demos. The second wave is deployment. If you are not using AI tools to augment your work, you are falling behind.
The competitive landscape shifts. Small companies can now access capabilities that were previously enterprise-only. The playing field levels, then tilts toward whoever adopts fastest.
The Bottom Line
Weiss Ratings AI is Second Wind report is marketing. But the underlying trend is real.
The AI market is maturing. The infrastructure is built. Now comes the hard part: making it profitable.
Some companies will figure it out. Many will not. The winners of the second wave will not be the ones with the biggest models or the most funding. They will be the ones who solve real problems for real customers who pay real money.
That is not as exciting as AI will change everything. But it is how technology actually transforms the world.
One profitable use case at a time.
Related Reading
NVIDIA is Physical AI Play: Teaching Robots to Think, Talk, and Actually DO Things — How NVIDIA is building the bridge from AI training to physical deployment
The 5 Billion Tech Deal Spree: How AI Infrastructure Became the New Arms Race — The first wave of AI infrastructure investment and what comes next
The 3 Billion Bet: Why Microsoft Is Building Its Own AI Stack — Inside Microsoft is infrastructure play and what it means for the market
The UK Just Told Microsoft Its AI Strategy Needs Regulatory Supervision — How regulation is shaping the second wave of AI deployment
Foxconn is .6 Billion Quarter Proves AI Infrastructure Is the Real Money — The economics of building AI infrastructure at scale
