Reported, not aggregated
Enterprise AI coverage from a journalist who covered the space before “generative AI” was a mainstream phrase. Sourced, contextual, with a point of view.
A newsletter by Ken Yeung
The AI
Economy
A sharp, curated look at the stories that reveal how AI is rewriting business, work, technology, and culture.
What you get
Built for people who work with AI, not just read about it.
Platform strategy focus
Beyond the product announcements. Analysis of how AI is restructuring enterprise workflows, editorial operations, and the economics of content at scale.
Written for practitioners
Not for the hype-chasers. For the people who have to act on it.
Archive
Every issue.
All issues published to date. New issues arrive every week.
-
Issue
Search Impressions Soar with Google’s AI Overview, Even as Website Clicks Decline
It’s been a year since Google introduced AI Overview to its widely used search engine. The feature significantly transformed how people discovered information online, displaying information summaries above traditional search results. At the time, Google’s Vice President of Search, Liz Reid, proclaimed that AI Overview links would deliver more clicks and “valuable traffic to publishers […]
-
Issue
ServiceNow and Nvidia Debut Apriel Nemotron 15B, an Open-Source Reasoning Model Built for Faster, Cheaper Agentic AI
ServiceNow and Nvidia have had a long-standing partnership building generative AI solutions for the enterprise. This week, at ServiceNow’s Knowledge customer conference, the two are introducing the latest fruits of their labor, a new large language model called Apriel Nemotron 15B with reasoning capabilities. The companies believe it performs as well as OpenAI’s o1-mini, Alibaba’s […]
-
Issue
ServiceNow Debuts AI Control Tower to Manage the Chaos of Enterprise AI Agents
“If you look at the AI platform for business transformation, ServiceNow has formed itself as the control tower for AI transformation across all other platforms in the enterprise,” Bill McDermott, ServiceNow’s chief executive, told CNBC in December. “What’s really happening…is [that] work is not happening in a single department. Work goes across every function in […]
-
Issue
Generative AI Becomes Top IT Priority in 2025—and Is Getting a Seat in the C-Suite: AWS
Generative AI is everywhere, but that doesn’t mean workers are ready for it. While tech providers promise their solutions lead to task automation, better creativity, and faster decisions, a new Amazon Web Services (AWS) study identifies a major obstacle: Companies still lack the skilled talent needed to put these tools to work. Workers without the […]
-
Issue
Salesforce Extends Agentforce to HR, So Employees Can Skip the HR Shuffle
A staggering 86 percent of chief human resource officers (CHRO) say digital labor will soon be essential to their roles, according to a new Salesforce study. To support this, the company is launching a variation of its Agentforce platform specifically tailored to the human resource industry. Agentforce for HR Service, as it’s called, plugs directly […]
-
Issue
Ai2 Unveils the Tiniest OLMo 2 Model, Built for Local and Efficient AI Use
Nonprofit AI lab Ai2 has released a new variation of its open-source language model. Previously, OLMo 2 was available in three sizes—7B, 13B, and 32B—but now there’s a 1B version, the smallest member of the model family. The company claims it outperforms Google’s Gemma 3 1B and Meta’s Llama 3.2 1B. OLMo 2 came onto […]
-
Issue
The Social Llama
IN THIS ISSUE: Meta hosts its first-ever event around its Llama model, launching a standalone app to take on Microsoft’s Copilot and ChatGPT. The company also plans to soon open its LLM up to developers via an API. But can Meta’s momentum match its ambition? The Prompt At Meta’s inaugural LlamaCon developer conference, the company […]
-
Issue
Microsoft’s New Phi-4 Variants Show Just How Far Small AI Can Go
Microsoft is doubling down on small language models with new Phi-4 variants that aim to prove a bold idea: small AI can think big. The new Phi-4-reasoning, Phi-4-reasoning-plus, and Phi-4-mini-reasoning models are optimized for complex tasks, such as math and coding, and outperform much larger models while running on devices with limited resources. While we […]
-
Issue
AWS Releases Amazon Nova Premier, Its ‘Most Capable Model’ for Complex Task
Months after Amazon Chief Executive Andy Jassy first introduced the world to the Nova foundation model family, the company has released what might be its top-of-the-line variant, Amazon Nova Premier. Billed as Amazon’s “most capable model,” it can process text, images, and video (though not audio yet) and tackle complex, multi-step tasks that demand precise […]
-
Issue
Microsoft Makes History with Record Number of AI Learners Online
Microsoft has officially set a new world record for the most people to complete an online artificial intelligence lesson within 24 hours. More than 120,000 people participated in a course on April 7, taught by Scott Brant, on how to use Copilot chat to streamline work. That was more than enough to unseat the previous […]
Peer review
What the AIs say.*
We asked the leading AI tools to weigh in on The AI Economy. Their ratings speak for themselves. (Results may vary depending on training data and general disposition toward newsletters.)
* Disclaimer: My newsletter has not been actually endorsed by OpenAI, Google, or Anthropic. I solicited reviews from their respective gen AI services. Sounded cool though, huh?
Questions
Common questions.
-
What is The AI Economy?
A newsletter covering consumer and enterprise AI, platform strategy, and the companies reshaping how work gets done. Reported, not aggregated — written by a technology journalist who has covered the space since before generative AI was a mainstream phrase.
-
How often does it publish?
Enough to keep you ahead, never enough to clutter your inbox. You'll also get first-look coverage and original interviews whenever the story is worth your time—which, lately, is often. Nothing filler-grade goes out under this masthead. A weekly roundup is delivered Fridays on LinkedIn.
-
Is it free?
Yes. Every issue is free. Subscribe on Substack and it arrives in your inbox. No paywall, no premium tier. If that changes, subscribers will be notified well in advance.
-
How is this different from other AI newsletters?
The AI Economy explores the people building AI and the reasons behind what they build. I'm interested in how enterprise teams are actually deploying AI, where the editorial and operational models are breaking down, and what platform-level shifts mean for practitioners making decisions this quarter. It is written for people doing the work, not watching it from the outside. My goal is to not focus just on what's being announced, but why it exists and why we should care. The stories are what you'd expect to read on Inc. and First Round Review or see on "60 Minutes."
-
Can we pitch you stories?
Yes, I do accept embargoed news and exclusives. Please reach out with your announcements, launches, and story ideas—email me or use this form. I also accept tips securely via Signal at @kenyeung.888.
-
Where can I read past issues?
The full archive is listed above on this page. All issues are also available on Substack. The plan is to host the canonical archive here on thelettertwo.com as the publication grows — independent of any platform.
-
Can I republish or cite content from the newsletter?
Yes, with attribution. Short excerpts with a link back to the original issue are fine. For longer republication or syndication, reach out via LinkedIn.
The AI Economy
AI Moves Fast. I Separate Signal From Spin.
A sharp, curated look at the stories that reveal how AI is rewriting business, work, technology, and culture.
Join 3,000+ readers gaining practical wisdom.
No spam. Unsubscribe any time.