The Killer Feature (That Wins the AI wars?)
Gemini's Personal Intelligence paves the road for Life OS.
This could be the killer feature that wins the LLM wars.
Yesterday, Google rolled out “Personal Intelligence” for Gemini Pro and Ultra users in the US. The feature (which users have to opt in to) allows Gemini to reason across your entire Google ecosystem (Gmail, Photos, YouTube, Search history) to connect the dots that other models can’t see.
Example: You ask about winter tires. Instead of generic recommendations, Gemini finds your car model from an old Gmail receipt, checks your Photos for the tire rim size, then searches the web for the best winter tires for that specific setup.
This is Life OS. The holy grail of all tech apps. And for a good number of those who are on Google Workspace, it’s Work+Life OS.
Just as Google indexed the open web, they can now index your personal information history (through “context packing”) to make it relevant, contextual, and useful. There might feel be some discomfort at first at the “creepiness” of the idea, but if history tells us anything, extreme convenience will likely win out.
This turns decades of Google’s data infrastructure into an unassailable advantage.
Paired with Universal Commerce Protocol, this feels like the Google universe upgrading from managing your information to managing your decisions. From suggesting the right tires to buying the right tire you need.
Why This Matters
The LLM wars may not be won by the model with the highest benchmark scores. They may be won by the model that knows you best.
Google just made that case. And they’re the only ones who can.
Next, I anticipate that Google will aggressively make moves to build out their developer and enterprise partnerships on top of Gemini / Personal Intelligence / Work + Life OS.
If they open the gates to the TPU/Gemini stack for builders to create "Context Apps" on top of this personal data, the lock-in becomes permanent.
The Download
News that mattered this week
SoftBank and OpenAI commit $1bn to SB Energy for US AI data centre push: SoftBank Group and OpenAI are making a significant bet on the future of AI infrastructure, committing a combined $1bn investment into SB Energy as they partner to develop next-generation AI data centres in the US. The investment builds upon a much broader, previously pledged $500bn commitment made public at the White House earlier this year.
Mark Zuckerberg says Meta is launching its own AI infrastructure initiative: Meta CEO Mark Zuckerberg announced the launch of Meta Compute, a new initiative designed to bolster the tech giant’s AI infrastructure. Zuckerberg said the company intended to drastically expand its energy footprint in the coming years.
Anthropic launches Cowork, a Claude Desktop agent that works in your files – no coding required: Anthropic released Cowork, a new AI agent capability that extends the power of its wildly successful Claude Code tool to non-technical users.
Elon Musk’s xAI partners with Pentagon for Grok AI military integration: Elon Musk’s artificial intelligence venture, xAI, has secured a pivotal agreement with the U.S. Department of War, marking a significant expansion of AI’s role in military operations. This deal integrates xAI’s Grok models into the Pentagon’s GenAI.mil platform, enabling real-time data analysis and enhanced decision-making for defense personnel.