Wedbush’s Dan Ives: AI Revolution Is Only in Third Inning. Here’s Why Nvidia and U.S. Chip Stocks Have Years of Upside Left.

Wedbush’s Dan Ives went on The Pomp Podcast with Anthony Pompliano and laid out a thesis that should reframe how you think about every AI position in your portfolio. His line: “This is 3rd inning, 1 out in a 9-inning game.” If you accept that framing, the implication for chip stocks, hyperscalers, and even the utilities powering the buildout is straightforward. Most of the game is still ahead.

I’ve owned NVIDIA for over 15 years and Alphabet since April 2012, so I’ve watched several “peak AI” calls come and go. Ives’ argument is that we’re nowhere near a peak. He frames it as a “multi-year bull market relative to what’s happening in tech”, with the U.S. owning the parts of the stack where the value actually accrues: “the models, the software, software and the chips.”

Nvidia: The Godfather Trade

Ives calls Jensen Huang the “godfather of AI”, and the numbers behind NVIDIA (NASDAQ:NVDA | NVDA Price Prediction) keep validating that label. Q1 FY27 revenue hit $81.615 billion, up 85% year over year, with Data Center revenue of $75.246 billion growing 92%. Networking inside that segment grew 199%. Huang described the moment in the 8-K filing as “the largest infrastructure expansion in human history.”

Management is putting capital where the conviction is. The board authorized an additional $80.0 billion in repurchases and raised the dividend from $0.01 to $0.25 a share. Supply commitments now sit at $119.0 billion. Ives’ competitive read: “a third-rate Nvidia chip is a year and a half, 2 years ahead of Huawei.” That’s his claim, not an independently verified benchmark, but it’s the lens he’s using.

Shares are up 62% over the past year and 15% year to date. Polymarket assigns a 71.5% probability NVDA finishes higher on May 26.

The Software and Cloud Layer

Alphabet (NASDAQ:GOOGL) posted Q1 FY26 revenue of $109.90 billion, with Google Cloud growing 63% and backlog nearly doubling to over $460 billion. Microsoft (NASDAQ:MSFT) reported its AI business at a $37 billion annual run rate, up 123% year over year, with commercial remaining performance obligation of $627 billion. Polymarket gives Microsoft a 70% probability of out-valuing OpenAI and Anthropic combined by year-end 2027.

AMD: The Second Chip Story

Advanced Micro Devices (NASDAQ:AMD) is the loudest counterexample to the “Nvidia owns everything” reflex. Q1 FY26 revenue came in at $10.25 billion, with Data Center up 57%. Meta committed to up to 6 GW of AMD Instinct GPU deployment. Shares are up 322% over the past year and 118% year to date. If you take Ives’ framing seriously, multiple credible chip vendors can win because the pie is still being baked.

Utilities as a Derivative of AI

Ives’ most underappreciated point on Pomp was that utilities become “a derivative of AI” with expanding multiples as data center power demand scales. NextEra Energy (NYSE:NEE) added 4 GW to its renewables backlog in Q1 and was selected to build 9.5 GW of gas-fired generation under the U.S.-Japan trade deal. Management is guiding 8%+ EPS CAGR through 2032. The stock is up 36% over the past year.

The Losers and the Patience Trade

Ives’ framing for companies that resist AI is blunt: “typewriters” or “horses.” On the geopolitics, he believes when “Trump and Xi meet, they recognize I need you, you need me”, which keeps a hard decoupling off the table. China leads in robotics deployment and power generation. The U.S. captures the margin where the models, software, and chips live.

The takeaway from a third-inning game is patience. If Ives is right about the runway, the costly mistake at this stage is selling the winners early. The second mistake is assuming the trade ends at the obvious AI tickers. The power producers, the networking suppliers, and the second-source chip designers all sit inside the same buildout. Whether you size positions accordingly depends on how much of his thesis you believe.