$Apple(AAPL)$ The recent report on Google's Frozen v2 architecture really underscores the value of owning your own infrastructure.
While some are focused on trying to fit basic LLMs into limited mobile memory with software tricks, Google is taking a different path by directly integrating the Gemini neural architecture into its custom server silicon.
The potential efficiency gains are significant. By embedding the model blueprint into the hardware, Google is aiming for a 6x to 10x improvement in power and cost efficiency for inference tasks compared to standard TPUs.
This gets to a core economic issue. As spatial computing and continuous AI agents become more prevalent, the cost per inference token becomes a critical competitive advantage. If Google can serve tokens at a fraction of the cost of others, it could reshape the economics for competitors who rely on a more consumer-focused, asset-light model.
There seems to be a notable valuation gap in the market right now. One company is priced around 40x P/E, with assumptions of perpetual hardware growth and free AI backend support. Meanwhile, Google trades at about 19x P/E, with narratives often focused on challenges in search and capital expenditure, overlooking its infrastructure assets like custom silicon, cloud capacity, and its operating system for extended reality.
When the market reassesses these positions—recognizing one as a hardware distributor potentially facing infrastructure costs and the other as owning the underlying tech stack—the adjustment in valuations could be significant. Analyst upgrades for the high-flyer might continue, but the fundamental picture looks different.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
