This wasn't a year where Google released AI features. It was the year AI stopped being something Google releases — and became what Google is. The next 18 months of AI work won't look like better chat windows. It'll look like agents finishing work before you ask.
We’ve agreed to a partnership with @SpaceX that will substantially increase our compute capacity.
This, along with our other recent compute deals, means that we’ve been able to increase our usage limits for Claude Code and the Claude API.
Most people are adapting to AI.
A smaller group is reinventing around it.
Adapting = using AI to do the same work faster.
Reinventing = looking at your day and asking what can genuinely be outsourced — and what's left when it is.
The gap between them is growing fast.
Hot take: debating which AI model to use is the most productive-feeling unproductive thing happening in marketing right now.
The model isn't the variable. How you use it is.
Everyone's asking "which model should we use?"
That's the wrong question. Here's what actually separates the teams seeing results from the ones still waiting 🧵
Ran the same brief through three different models today. The outputs were far more similar than different.
The gap isn't the model. It's how you brief it.
The teams winning with AI aren't using a better model than you.
They're using AI as a thinking partner, not a content machine. They feed it context. And they've integrated it into how work actually gets done.
Model choice is the last thing that matters.
The "AI replaces jobs" debate is binary and that's the problem. At stable performance? Yes, your team can do more with less. Scaling into a $100M account? You're hiring across every department. Context matters more than the headline.
I'd rather hire one expensive senior person who's supercharged on AI than two juniors who aren't using it. The hiring equation has changed and most companies haven't updated the formula.
I keep hearing "we need more people" from teams that aren't growing. Same clients. Same workload. Stable year-over-year. That conversation looks very different today than it did three years ago.
The teams winning with AI aren't using a better model than you.
They're using AI as a thinking partner, not a content machine. They feed it context. And they've integrated it into how work actually gets done.
Model choice is the last thing that matters.
AI replaces jobs" is missing the biggest caveat: it assumes baseline performance. Growing exponentially? You're still hiring. The real question isn't whether AI replaces people. It's whether you're at baseline or in growth mode
Hot take: the AI infrastructure conversation is a marketing campaign by cloud vendors. For most individual builders, the real setup is a laptop, WiFi, and a CLI. Everything else scales with revenue — it doesn’t need to lead it.
Enterprise reports quote $25K-$300K+ for agent development. My five agents are running on a laptop that was literally furniture three months ago. The infrastructure narrative is keeping individual builders on the sidelines.