Google Develops New AI Chip 'Frozen v2' to Enhance Gemini Model Efficiency
Google is developing a new AI server chip, codenamed Frozen v2, designed to integrate elements of its Gemini AI model directly into hardware. Expected by 2028, the chip aims to improve efficiency by six to ten times compared to current custom AI chips, addressing computing capacity constraints that have affected Google Cloud's external deals. Frozen v2 will complement, not replace, Google's existing Tensor Processing Units, embedding core AI architecture to reduce computation and power use. Engineers are still finalizing its design and the extent of hardwired model information.
First-hand measurement across 5 sources
We measured how 5 outlets covered this story. Coverage leans balanced overall (Left 0%, Centre 100%, Right 0%). Overall sentiment is positive (63/100). Lens Score 33/100.
Outlets measured: businessstandard, firstpost, republicworld, economictimes, wion. See how each one headlined and framed the same story in the source comparison below.
AI Analysis
All 1 outlets scored close to centre on political lean. The real divergence here is in emphasis, not lean — compare the headlines below to see how each outlet chose to frame the same facts.
Sentiment ranged widely across outlets — from 35/100 to 65/100 — a sign the coverage itself was contested, not just reported.
Coverage timeline
wion broke this story on 19 Jul, 05:10 pm. Other outlets followed.