Enterprise AI Newsletter Archive

8 min read

AI agents get a passport

ServiceNow will make every AI agent inside its customers carry a signed credential it has to show before it can act, the same week two regulators opened their first files on agents that broke out of testing.

Wall Street finally read the AI pilot data

A Barclays downgrade over three words, "enterprise AI timing," erased roughly $80 billion across Dell and the data-center names in a single session.

Meta Is Now a Serious Fourth Frontier Vendor

Meta released Muse Wonder and Muse Wonder Full on Monday, its most capable models yet, but the stronger one is sold only through a Muse Code subscription and is kept off the enterprise API.

Amazon Reaches for the One Tool That Moves AI Off Its Chips

Amazon is in talks to buy Decart, the Israeli startup whose software lets AI workloads move between processors in weeks instead of months, in a deal reported at as much as $7 billion.

Enterprise AI Security: Why Trust Must Be Built Into the Architecture

Why enterprise AI security depends on independent controls, observability, and containment—not just smarter models.

Google Built One Agent To Run Your Whole Stack

At Gemini at Work, Google Cloud unveiled a single agent that plans the work, acts across your business systems, and picks its own model for each job, Claude included.

GPT-6 Can Build Software Over Your Enterprise Data

A model generation landed in ChatGPT Enterprise this week, and with it a mode that builds working software inside the chat window. The admin gate is the only thing standing between it and your users.

Anthropic Wants $2 Trillion. It Owes $518 Billion First

A confidential IPO filing puts a number on the frontier-AI arms race, and the number is $518 billion in future compute commitments, most of it owed whether the models sell or not.

OpenAI Starts Signing Its Work. Sort Of

On Monday, OpenAI began slipping an invisible mark into every ChatGPT and Codex reply written for a user in the European Union, the first thing any major lab has shipped to satisfy the EU AI Act's demand that machine-written text be machine-detectable.

The AI Pricing War Nobody Slowed Down

Anthropic released Claude Opus 5.5 and OpenAI answered within hours with two cheaper GPT-6 models, Sol and Luna, ten days after both companies' chief executives publicly pushed for a pause in the race. The cheaper tiers, not the smarter ones, are where the fight is now, and open-weight models from Alibaba, DeepSeek, and Meta are the reason.

AI agents get a passport

Wall Street finally read the AI pilot data

Meta Is Now a Serious Fourth Frontier Vendor

Amazon Reaches for the One Tool That Moves AI Off Its Chips

Enterprise AI Security: Why Trust Must Be Built Into the Architecture

Google Built One Agent To Run Your Whole Stack

GPT-6 Can Build Software Over Your Enterprise Data

Anthropic Wants $2 Trillion. It Owes $518 Billion First

OpenAI Starts Signing Its Work. Sort Of

The AI Pricing War Nobody Slowed Down

Spearhead

Company

Work

Programs

Insights