DAILY INTELLIGENCE BRIEF

The Vault - AI Edition

Saturday, June 13, 2026
15 stories
14,398 words
67 min read

Business

5
01

Anthropic Overtakes OpenAI in U.S. Business Adoption for the First Time

The numbers had been converging for months. In April, they finally crossed: Anthropic's share of U.S. business adoption pulled ahead of OpenAI's.

4 minRead →
02

EngineAI Files Confidentially for a Hong Kong IPO as China's Humanoid Robot Race Accelerates

In China's fast-moving humanoid robot industry, Shenzhen-based EngineAI has gone from founding to factory floor to IPO candidate in under three years.

4 minRead →
03

Anthropic and OpenAI's Revenue-Accounting Split Becomes an IPO Flashpoint

Two of Silicon Valley's most valuable private companies are racing toward public markets — and can't agree on how to count their revenue.

4 minRead →
04

Anthropic Is Blindsiding Its Own Business Partners, The Information Reports

A new report from The Information paints an uncomfortable portrait of Anthropic's partner relations: a pattern of surprise competitive launches.

4 minRead →
05

IBM and ServiceNow Expand Their Alliance to Run Agentic AI on Enterprises' Legacy Systems

For the world's largest enterprises, the biggest obstacle to AI at scale is not a shortage of models — it is decades of legacy infrastructure.

5 minRead →

Llms Genai

3
06

A Leaked 'GPT-5.6' Checkpoint Is Circulating — Stronger Vision, Bigger Context, No Official Word

A checkpoint identified as GPT-5.6 'kindle-alpha' began circulating in developer communities in early June, surfacing through Codex testing paths.

4 minRead →
07

Alibaba's Qwen 3.7 Plus Folds Vision, Language and Coding Into One Multimodal Model

While Microsoft Build 2026 commanded the stage in Redmond, Alibaba quietly dropped one of the most economically consequential model releases.

4 minRead →
08

Claude Code Gains a '/fork' Command and a New CLI for Branching AI Coding Sessions

Anthropic has given its coding tool a feature that sounds small and changes the workflow more than its size suggests: session forking.

4 minRead →

Research

4
09

D-I-TASSER Marries AI With Physics to Predict Tough Protein Structures More Accurately

For years, researchers hailed AlphaFold as the answer to biology's protein-folding challenge. A new hybrid tool targets where it still struggles.

4 minRead →
10

Google Researchers Ask When Multi-Agent AI Actually Helps — and When It Backfires

The AI industry has been racing to stack agents. New research from Google challenges that bet with hard numbers about when it fails.

4 minRead →
11

Study Finds Large Reasoning Models Can Act as Autonomous Jailbreak Agents

A new peer-reviewed study documents something AI-safety researchers have long feared: a capable reasoning model that can jailbreak others on its own.

5 minRead →
12

Robin, an AI 'Scientist' Agent, Runs the Drug-Discovery Loop From Hypothesis to Analysis

AI in drug discovery has mostly meant point solutions. Robin tries to automate the whole loop — and produced its first candidate molecule.

4 minRead →

Policy

3
13

CNN Sues Perplexity as the Publisher-vs-AI Copyright War Widens to Nine Plaintiffs

The lawsuit the AI industry had been bracing for arrived on May 28, when CNN filed a federal complaint against the 'answer engine.'

4 minRead →
14

Europe Tightens the Screws on AI Training Data: A 250M Euro Google Fine and a German Order Against OpenAI

Europe is not waiting for legal consensus. A fine against Google and a court order against OpenAI mark a shift from warnings to enforcement.

4 minRead →
15

The EU AI Act's August Deadline Looms: What General-Purpose AI Providers Must Soon Disclose

Seven weeks out, on August 2, 2026, the EU AI Act's enforcement regime clicks into its next gear for general-purpose model providers.

5 minRead →