DAILY INTELLIGENCE BRIEF

The Vault — AI Edition

Wednesday, July 1, 2026
16 stories
15,504 words
75 min read

Business

5
01

California Signs Landmark Anthropic Deal, Bringing Claude to All State Agencies at Half Price

When Gov. Gavin Newsom stepped to the microphone in Sacramento on June 29, he did not announce a pilot or another task force.

5 minRead →
02

National Grid Bets $1.75 Billion on AI Data Center Power With Joulent Stake

Britain's National Grid, one of the oldest and most conservative names in global utilities, is placing a $1.75 billion wager.

5 minRead →
03

Bloom Energy and Brookfield Expand AI Power Partnership Fivefold to $25 Billion

When Bloom Energy and Brookfield first agreed to build and finance fuel-cell power for AI data centers, the number was $5 billion.

5 minRead →
04

Canadian Defense Startup Dominion Dynamics Raises $100 Million in Record Series A

Ottawa-based Dominion Dynamics has closed a $139-million CAD Series A, described as Canada's largest defense-tech round.

4 minRead →
05

Verkada Partners With Nvidia to Push Physical AI Into Buildings and Security Systems

Verkada, the cloud-based physical security company whose cameras and sensors watch over 30,000 organizations, is going deeper on AI.

5 minRead →

Llms Genai

4
06

Google Releases Gemini 3.1 Flash Image and Gemini 3 Pro Image as It Fills the Lineup Awaiting 3.5 Pro

Google shipped two image models on Tuesday, and neither one was the model everybody is waiting for.

5 minRead →
07

OpenAI to Run GPT-5.6 Sol on Cerebras at 750 Tokens a Second, Chasing Real-Time AI

When OpenAI unveiled its GPT-5.6 model family on June 26, the headline benchmark scores were not the most consequential number.

5 minRead →
08

xAI's Grok 4.3 Arrives on Amazon Bedrock as the Cheapest Major-Lab Model on AWS

When xAI slipped Grok 4.3 onto Amazon Bedrock in mid-June, it did not arrive as a headline-grabbing frontier launch.

4 minRead →
09

Anthropic's Claude Goes Generally Available in Microsoft Foundry on Nvidia GB300 Blackwell Ultra

Enterprise AI rarely stalls on model quality. It stalls on procurement, governance sign-offs and data residency.

4 minRead →

Research

3
10

'Squidbleed': Claude Mythos Uncovers a 29-Year-Old Memory Leak in the Squid Proxy

For 29 years, a single line of C code in one of the internet's most widely deployed proxy servers quietly leaked secrets.

5 minRead →
11

KAIST's VOTP Teaches Robots Human Judgment From a Handful of Preference Videos

Teaching a robot what humans actually want has long been one of the most expensive problems in artificial intelligence.

4 minRead →
12

Anthropic Launches Claude Science, a Dedicated AI App for Drug Discovery and Biology

When Dario Amodei stood up in San Francisco on Tuesday, he did not pitch Anthropic's newest product as a chatbot with a lab coat.

5 minRead →

Policy

4
13

Five Eyes Issue Rare Joint Warning: AI-Powered Cyberattacks Are Months Away, Not Years

Spy agencies do not usually agree on the wording of a press release, let alone on a deadline.

5 minRead →
14

UN Scientific Panel Warns Unchecked AI Progress Could Create Catastrophic Risks

The world's first fully independent, UN-backed scientific body on artificial intelligence delivered its verdict on Wednesday.

5 minRead →
15

EU and Apple Clash Over Siri AI Delay as the Digital Markets Act Meets Generative AI

The most consequential meeting in the generative-AI industry this week was not a product launch. It was a video call.

4 minRead →
16

Bank of England Official Says AI Trading Systems May Need 'Kill Switches'

Speaking to a room of central bankers in Sintra, Portugal, one of the Bank of England's most senior officials floated a kill switch.

5 minRead →