The move also comes as investors await Alphabet’s second-quarter earnings on July 22.Google Develops AI Chip To Boost EfficiencyGoogle is developing an internally designed server chip code-named “Frozen v2,” according to The Information. The chip is intended to embed parts of Gemini’s architecture directly into the hardware, reducing the amount of computation and data movement required to generate AI responses.Engineers believe the chip could deliver six to 10 times more AI tokens per unit of power than Google’s latest custom AI chips. The company is targeting deployment by 2028, although the design remains under development, CNBC reported Monday.The company is working on a new server chip that could help Google run Gemini models more efficiently and ease pressure from rising AI computing demand.Custom Silicon Could Help Relieve AI Capacity ConstraintsFrozen v2 would complement, rather than replace, Google’s existing tensor processing units. The project is aimed at improving AI inference efficiency and easing growing pressure on the company’s computing infrastructure.According to the reports, AI capacity constraints have created internal bottlenecks and forced Google Cloud to limit some external customer demand, making more efficient custom chips increasingly important as Gemini usage expands.Alphabet Technical AnalysisFrom a longer-term trend view, GOOG is still in an uptrend over the past year (up 86.60%) and is trading 10.9% above its 200-day SMA ($321.52), which keeps the bigger-picture bias constructive.Nearer-term, it’s more mixed: the stock is 0.9% above the 20-day SMA ($353.44) but trading 2.9% below the 50-day SMA ($367.20), and the 20-day SMA sitting below the 50-day SMA is a bearish short-term alignment.Earnings & Analyst OutlookThe countdown is on: Alphabet is set to report earnings on July 22, 2026 (confirmed).