The current technological landscape bears a striking resemblance to the market dynamics of 1997-1998, prompting seasoned Wall Street analysts to anticipate a sustained AI bull market despite looming volatility. On May 11, Dan Niles, founder of Niles Investment Management and a veteran chip analyst from the Internet bubble era, articulated this thesis on The Master Investor Podcast. He posits that while the AI rally is not exhausted, a significant correction is probable around early 2027, urging investors to prepare immediately. This outlook aligns with a JPMorgan survey of 56 global investors, where 54% anticipate a U.S. stock market pullback exceeding 30% within the next 12 months, and 45% specifically pinpoint 2027 as the likely inflection point.

Niles draws a precise historical parallel between the current AI infrastructure build-out and the Internet era of the late 1990s. With ChatGPT launching in late 2022, the sector has entered its fourth year of development, mirroring the timeline when Netscape appeared in 1994 and the market matured in 1997-1998. During that period, macroeconomic shocks such as the 1997 Thai currency crisis and the 1998 Russian bond defaults caused the S&P 500 to drop 11% and 19% respectively within a year, yet the index still posted annual gains of 31% and 27%. Data compiled by Woofun AI shows that similar to the past, current infrastructure development provides a supportive backdrop where macro shocks, including oil price spikes from the Iran war, are viewed as temporary buying opportunities rather than structural threats.

The core catalyst for this year's market trajectory is the emergence of Agentic AI, a paradigm shift from simple dialogue-based interactions to autonomous task execution. Unlike previous models that merely answered queries, Agentic AI can orchestrate complex workflows, such as retrieving data from multiple sources like BBC, Bloomberg, and CNBC to generate an Excel spreadsheet. This operational complexity demands 10 to 100 times more computing power and tokens than dialogue-based AI. Evidence of this surge is visible in token consumption metrics: two months prior to the January 30, 2026 release of Open Claw, token growth was approximately 20%, but it surged to over 120% in the subsequent two months. This qualitative leap has forced large cloud service providers to revise capital expenditure expectations for 2026 upward from an initial 30% to 60% after Q1 reports, and further to 70% following the latest financial disclosures.

This architectural shift is quietly reshaping the competitive landscape of AI hardware, favoring CPUs over the GPU dominance of the previous cycle. While GPUs excel at the repetitive tasks of model training and dialogue generation, Agentic AI requires the simultaneous scheduling and orchestration of multiple applications, a function where CPUs hold a distinct advantage. Niles projects the hardware ratio will shift from the historical 8 GPUs per 1 CPU to a near 1:1 balance, potentially benefiting Intel and AMD while pressuring Nvidia's valuation. Woofun AI notes that despite short-term overbidding in semiconductor ETFs, which have risen approximately 70% this year, the long-term demand logic driven by Agentic AI remains robust, justifying a strategic tolerance for potential 15-20% short-term price corrections in chip stocks.

Looking toward the next cycle, Niles forecasts that the growth rates achieved in early 2026 will face a higher base effect by early 2027, naturally slowing momentum and triggering a market correction of 30% to 50% from current highs. This prediction echoes the 2022 decline of the "Magnificent 7" tech companies, which fell an average of 46%, though that event was driven by pandemic-related retreats rather than the scale of the current AI boom. A critical structural risk involves OpenAI and Anthropic, which together account for roughly half of the order backlog for major cloud providers. Their combined annual revenue has exploded from approximately $7 billion at the start of 2025 to nearly $70 billion today, with Anthropic contributing $45 billion and OpenAI $24 billion.

However, the sustainability of OpenAI's claimed $1.4 trillion capital expenditure plan over eight years raises liquidity concerns, especially as imminent IPOs for OpenAI, SpaceX, and Anthropic could force fund managers to sell other assets to meet trillion-dollar valuations.

Macro indicators further complicate the outlook, with oil prices rising approximately 60% this year and 10-year and 30-year U.S. Treasury yields reaching yearly highs. Historically, continuous oil price increases preceded 10 out of the last 12 recessions. If oil prices stabilize around $90 per barrel for one to two quarters, inflationary pressure could erode consumer purchasing power, a risk highlighted by McDonald's recent report of same-store sales missing expectations due to low-income consumer strain. Woofun AI analysis suggests that while Federal Reserve Chairman Kevin Warsh views AI as a deflationary force and may cut rates, the convergence of rising bond yields, high oil prices, and record stock valuations creates a fragile equilibrium where a correction in any single variable could trigger significant market turmoil.

Recent financial reports from technology giants validate the divergence in strategic positioning. Google Cloud demonstrated the strongest momentum, accelerating its year-on-year growth from 48% in Q4 to 63% in the most recent quarter, a 15 percentage point increase. AWS saw growth rise from 24% to 28%, while Microsoft Cloud remained flat at 38% to 39%. Niles identifies Google as the clear winner for the next 3 to 5 years due to its complete technology stack, including the Gemini model, self-developed AI chips with over a decade of history, and an Android ecosystem covering more than 75% of global smartphones. Conversely, Meta faces headwinds with no public cloud platform, lagging ASIC chip development, and its first quarterly decline in Family of Apps users, signaling slowing advertising revenue drivers.

In conclusion, the market philosophy for the coming cycle is to remain nimble, holding firm beliefs loosely while monitoring oil prices, bond yields, and equity valuations daily. The immediate momentum is supported by Agentic AI and loose monetary policy, but the trajectory points toward a 30-50% correction by early 2027 as growth rates normalize against a higher base. Investors are advised to maintain significant cash reserves to navigate the potential liquidity shocks from upcoming mega-IPOs and macroeconomic adjustments, ready to deploy capital only when the resolution of these structural imbalances becomes clear.