New York, January 12, 2026, 11:02 EST
- Alphabet briefly soared to a $4 trillion valuation as investors responded to its revamped AI strategy
- Shares edged mostly flat in late morning trading, following a morning rally.
- Attention now turns to Google Cloud’s expansion, early takes on Gemini 3, and efforts to monetize proprietary AI chips
Alphabet surged past a $4 trillion market cap on Monday, highlighting the rapid shift in investor sentiment toward Google’s AI ambitions. 1
This shift matters now as Big Tech’s AI spending spree bumps up against a crucial question: who can convert their hardware and models into steady cash flow, and who’s merely leasing pricey compute power from others.
Alphabet shares dipped 0.3% to $327.47 in late morning trading. The stock has climbed roughly 65% over the last year and gained about 6% year-to-date in 2026.
The rally has propelled Alphabet closer to Nvidia, Microsoft, and Apple among the market’s elite, shedding its old image as just a mature ad company and stepping into the AI spotlight.
Alphabet has boosted that re-rating by shifting Google Cloud’s image from an internal expense to a growth driver. Reuters reported a 34% jump in Cloud revenue in Q3, with a backlog of signed but unrecognized sales contracts swelling to $155 billion.
Investors are zeroing in on reviews of Alphabet’s Gemini 3 model. Reuters reported it sparked strong reactions, ramping up the pressure on OpenAI following mixed responses to its recent GPT-5 launch.
Hardware is another key area. Google is starting to rent out its in-house AI chips to external clients, promoting its “Ironwood” TPU series as a more affordable option for running large models—especially during inference, where trained models deliver real-time answers. James Bradbury, head of compute at Anthropic, said in a Google Cloud blog post, “Ironwood’s improvements in both inference performance and training scalability will help us scale efficiently while maintaining the speed and reliability our customers expect.” 2
Some investors push the idea even further, saying Alphabet’s vertical integration could lead to what they call “near-zero marginal cost” AI. That means the expense of handling one extra query drops dramatically once the system is in place. “GOOG’s internal silicon … enable near-zero marginal AI inference costs,” noted Seeking Alpha contributor Esxeleryn Analytics in a Jan. 10 report. 3
On the chip front, Reuters cited a report saying Meta is considering a multibillion-dollar purchase of Alphabet’s chips for its data centres beginning in 2027. That move would pit Google head-to-head with Nvidia’s dominant GPU business.
The downside risks haven’t disappeared. AI assistants might drain revenue from search ads, and robotics continues to show that edge cases remain a thorny problem. In Phoenix, a Waymo robotaxi stopped on light rail tracks just as a train approached, forcing the passenger to jump out. “Something unexpected where the machine drove like a machine rather than a person,” said Arizona State University professor Andrew Maynard. 4
At this stage, the market views Monday’s valuation update as signaling that Alphabet’s AI narrative has lost its defensive edge. The upcoming challenges are more straightforward: can Cloud continue turning backlog into actual revenue, and will chip rentals evolve from a minor sideline into a genuine business segment?
