For the past few months, I've been writing a weekly AI newsletter for my team at the digital agency where I worked. It's called the dAIgest (I know, I know... but the pun was right there), and it started as a way to keep a team of designers, developers, and strategists up to speed on an industry that moves faster than any of us can individually track. The feedback has been good enough that I figured: why not share it more broadly? So here we are.
Welcome to the dAIgest: now on camhatesrunning.com, where it joins my other niche interest. If you've been feeling like AI news is overwhelming, contradictory, or hype-driven, this is for you. Let's catch you up.
Where Things Stand Right Now (May 2026)
If you haven't been paying close attention to the AI space and you're just tuning in: a lot has happened. Rather than bore you with a timeline, let me focus on the five things that define this moment.
1. The Money Is Staggering, and Getting Bigger
The four biggest tech companies (Amazon, Google, Meta, and Microsoft) are collectively spending roughly $700 billion on AI infrastructure this year. They're building massive data centers, buying specialized chips by the millions, and locking in energy contracts that rival small nations' power grids. For context, combined capital expenditures from these four companies exceeded $130 billion in Q1 2026 alone (Fortune), and the spending could surpass $700 billion for the full year, up sharply from about $410 billion last year (Bloomberg).
An NYU professor called it "the greatest capital misallocation in history." (Yahoo Finance). The companies themselves say they'd rather overbuild than get left behind. The truth is probably somewhere in between, and we won't know which for a few years.
What it means for you: If you own an S&P 500 index fund, you're already invested in this bet whether you realize it or not. The top 10 companies in the index (most of them AI-driven) now make up over 40% of total market cap. Whether we like it or not, the American economy is now riding on the bet of AI.
2. The AI Tools You Can Actually Use Are Getting Really Good
Forget the sci-fi stuff for a second. The AI tools regular people can use right now (Claude, ChatGPT, Gemini, Perplexity) have gotten dramatically better in a short period of time. The current generation of models can write competent code, analyze complex documents, help you plan a project, draft emails that don't sound robotic, and reason through multi-step problems. They're not perfect (more on that later), but they've crossed a threshold where they're genuinely useful for knowledge work, not just party tricks.
The most interesting recent development on this front: Perplexity launched something called "Personal Computer", an AI agent that runs on a dedicated Mac mini and works across your local files, email, and apps 24/7. It doesn't wait for you to ask it something: it monitors, executes, and carries work forward between sessions. Perplexity's enterprise version reportedly completed the equivalent of 3.25 years of work in four weeks. We're moving from "AI as a search box you type into" to "AI as a coworker that runs in the background."
What it means for you: If you tried ChatGPT a year ago and thought "meh," it's worth revisiting. The tools have improved substantially, and the new wave of AI agents that can actually do things on your computer (not just answer questions) is a genuine step change.
3. Most People Still Aren't Using It
Here's the number that should be getting more attention: a global survey found that more than 54% of workers bypassed their company's AI tools in the past 30 days (Fortune) and completed the work manually instead, and another 33% haven't used AI at all. That's roughly 80% of the workforce either avoiding or outright rejecting AI tools.
At the same time, Stanford's annual AI Index report found that people are adopting AI faster than they adopted the personal computer or the internet. AI is boosting productivity by 14% in customer service and 26% in software development (MIT) for those who do use it.
The paradox of the AI moment in 2026 is the backlash and the inevitability are not competing stories; they are the same story, told from different points on the adoption curve.
What it means for you: You're not behind if you haven't figured out AI yet. Most people haven't, and some are actively avoiding it. However, the gap between people who learn to use these tools effectively and those who don't is widening fast.
4. The Ethics and Politics Are Very, Very Real
Earlier this year, Anthropic (the company that makes Claude) refused to give the Pentagon unrestricted access to its technology. Specifically, they drew two lines: no fully autonomous weapons and no mass surveillance of Americans.
The Pentagon pushed. Anthropic held. Trump blacklisted the company as a national security risk - a designation normally reserved for foreign adversaries. Anthropic sued, and a federal judge blocked the ban, calling it "classic illegal First Amendment retaliation."
The twist? The military was already using Anthropic's AI tool through a Palantir integration to identify targets in the Iran conflict. The same model we use to prototype landing pages was simultaneously helping the military process satellite intelligence. A judge ruled the government can't punish a company for publicly disagreeing about how its technology should be used in war.
The consumer reaction was unexpected: Claude became the #1 downloaded app in the US during the controversy. Sometimes, standing for something is the best marketing you can do.
What it means for you: The AI tools you choose aren't neutral. The companies behind them have different values, different business models, and now, different relationships with the government.
5. The Job Market Is Shifting, But Not the Way Headlines Suggest
This is the question everyone actually cares about: is AI going to take my job? The honest answer: it's complicated.
Stanford's AI Index found that employment for software developers aged 22 to 25 has fallen nearly 20% since 2022. A third of organizations surveyed by McKinsey expect AI to shrink their workforce in the coming year, particularly in service, supply chain, and software engineering.
But it's not a simple "robots replace humans" story. Oracle just cut 20,000-30,000 jobs to redirect $8-10 billion toward AI. Block (the company behind Square) laid off 40% of its staff with the CEO citing AI as the reason. These aren't struggling companies, they're profitable firms making a strategic bet that they need fewer people and more machines.
At the same time, new job categories are emerging. McKinsey found that 8% of newly created job categories are AI-related. The people who understand how to work with AI (not just use it casually, but integrate it into workflows) are becoming increasingly valuable.
What it means for you: The safest position isn't "ignore AI" or "become an AI expert." It's "get comfortable enough with AI tools to use them well in your existing work." The people most at risk aren't those in any particular field - they're the ones who refuse to adapt.
Quick Hits: Stories Worth Knowing About
A few more things that happened recently that are worth having on your radar:
- Meta is recording employee keystrokes to train AI. Meta quietly deployed monitoring software on US employees' work laptops (TechCrunch) that captures keystrokes, mouse movements, clicks, and periodic screenshots to generate training data for AI agents. European employees are exempt because GDPR won't allow it. There's no opt-out for US workers. Meta laid off 8,000 employees the same week.
- YouTube can now detect deepfakes of celebrities. YouTube expanded its likeness detection technology to the entertainment industry, giving talent agencies and celebrities the ability to find and request removal of AI-generated content that uses their faces. It works like Content ID but for faces instead of copyrighted music (TechCrunch). Voice detection is reportedly coming later this year.
- OpenAI is spending $20B on non-Nvidia chips. OpenAI doubled its deal with chip startup Cerebras to over $20 billion over three years (TechCrunch), with equity warrants that could give OpenAI a 10% stake. The move signals a serious effort to break Nvidia's dominance over AI hardware. Cerebras is planning an IPO at a $35 billion valuation (Manila Times).
- SoftBank is spinning out a $100B AI robotics company. Called "Roze," it would focus on using robots to physically build AI data centers. Because apparently we've reached the point where we need robots to build the buildings that house the AI.
- Memory chip prices are spiking. AI data centers are consuming so much RAM that consumer electronics are paying the price. Mac minis with upgraded RAM are showing 4-5 month delivery estimates or are completely out of stock. If you've been putting off a computer purchase - or really, any product with some sort of computing capability, like cars, a new refrigerator, etc. - the shortage is real and it's AI's fault.
What I'll Be Covering Going Forward
The dAIgest will continue as a roughly weekly series here on the blog. Each issue will include:
- What happened: the biggest AI stories of the week, explained in plain language
- What it means: practical implications for regular people, not just tech insiders
- What to watch: what's coming next and why it matters
- One thing to think about: a closing thought that goes a little deeper
I'm not trying to be comprehensive; there are a hundred AI newsletters that try to cover everything. I'd rather cover fewer things well and help you actually understand them.
One Thing to Think About
Here's what strikes me most about this moment: the gap between what AI can do and what most people know it can do has never been wider.
We're living through what might be the fastest-moving technology shift in human history, and yet 80% of workers aren't using it. Companies are spending $700 billion on infrastructure while most of their employees manually do tasks that AI could handle. The most sophisticated AI model in the world is simultaneously helping someone plan a birthday party and helping the military plan airstrikes.
The AI story in 2026 isn't about the technology anymore. The technology works. The story is about adoption, values, economics, and power; who uses it, who controls it, who benefits, and who gets left behind. Those are human questions, not technical ones - and they're the ones I'm most interested in exploring with you.
Welcome to the dAIgest. Glad you're here.