Customer service bots that can’t answer what seem like basic questions. Executives who mandated AI everywhere and now own the cost overruns. Lawyers who got caught citing cases that don’t exist.

These stories are often treated as evidence that AI is overhyped and doomed to fail, maybe even that it should be discarded.

They’re better understood as evidence of learning. Of trial and error.

I’ve spent the past year figuring out how AI can help me. At various times, I’ve worked with AI to plan fitness regimens, shape marketing plans, and create elaborate automations on my phone to send morning briefings with the news and my to-do list.

Some of these uses I abandoned within a week. Some were almost right but required revision. And some, like using AI to help explore ideas, have worked from the start.

My misses were the only way to find out what worked for me. I couldn’t predict in advance where a tool would or wouldn’t fit my needs.

Likewise, we can’t derive from first principles how to use AI in customer service, or in management, or in law. The stakes might be larger for a business than for me, but the trial-and-error approach to learning is the same. Christopher Mims’ recent Wall Street Journal piece makes a complementary point: trial and error is how we learn where not to use AI.

We have to try things and watch what breaks.

Head up to a rooftop in Washington, DC on July Fourth and look around. There are fireworks in every direction.

None of these fireworks are official. Many of them are of dubious legality. All of them are joyfully celebrating the country‘s birthday.

The official fireworks on the Mall a few miles away are visible if you squint. They reflect a patriotism of pomp: Sousa marches, military flyovers, and elaborate displays planned months in advance.

The neighborhood fireworks are America. Informal. Self-organized. Ungoverned. Thumbing their noses at authority.

Here’s to another 250 years.

The New York Times reports on the recent Venezuela earthquakes:

Venezuela does not have a national early warning system of its own, but people with Android phones received alerts from Google’s Earthquake Alerts system, which can pull data from more than two billion phones equipped with built-in accelerometers. The same sensor that detects rotation on the screen can also sense vibrations from seismic waves.

This is recombination fueling innovation. Nobody put accelerometers in phones to create an earthquake-warning system. Instead, Google engineers stitched together built-in sensors, location data, push notifications, and seismology know-how to create something entirely new: a distributed, planet-scale earthquake-detection network.

Of course, this tech is possible only because the devices we carry with us are capturing and sharing, in real time, both our locations and the slightest vibrations.

Two labor economists argue that remote work explains a third of the recent rise in Americans’ mental distress.

Their proposed fix isn’t a full return to the old office norm. It’s “doses of in-person time with one another,” supported by employers who are more intentional about restoring some of the connection remote work has eroded.

They cite firms redesigning coffee spaces, rewarding “the too-often-invisible work of connecting teams,” and — more dubiously — turning managers into mentor-matchmakers.

I’m sympathetic to part of this. In-person work facilitates the informal knowledge-sharing and trust-building that makes work better. And the best offices can enable collaboration — Pixar famously designed its campus to increase chance encounters across teams.

But Pixar’s goal wasn’t to make employees less lonely. It was to improve collaboration. The friendships that followed were incidental. That distinction matters.

Employers should care whether people communicate well, trust one another, mentor junior colleagues, and share useful information across teams.

But that doesn’t mean employers should solve loneliness.

Work can produce meaningful relationships. But those relationships are a byproduct of work, not what work is for.

Your employer isn’t your camp counselor.

You just bought something. You’re excited.

This is the moment a brand has been working toward. After they spent money to acquire you, you’re in.

Now they need to reinforce to you that you made a great decision. Done right, this can make the next purchase feel like your idea.

Some brands get this right. Many don’t. Instead of welcoming you, they barrage you.

One recent shoe purchase triggered daily generic offers — pitching to me as if my last purchase never happened. Signing up for a streaming service earned me four emails on day one, and four more by noon on day two — almost begging me to watch something, anything.

It’s the difference between earning your loyalty and wearing you down until you open your wallet.