Using AI to Help With Photography in ways you wouldn't normally think.

Every time someone brings up AI and photography, the conversation goes the same place. Somebody gets nervous about fake sunsets and cloned-in eagles, somebody else brings up whether a photo edited with AI tools is still a "real" photograph, and forty-five minutes later nobody has said anything new. I get it. Then there is the whole other subject of using AI tools for editing and all of that. I have opinions on that stuff too. But that's not what this post is about.

What I don't hear many talks about is the boring, unglamorous ways AI actually saves me time before I ever pick up the camera. Not generating images. Not fixing skies. Just planning. Here are three things I actually use it for, no philosophy required.

Scouting Locations

The way everyone usually scouts is to drive somewhere based on a vague memory of a photo you saw once, get there, realize the light was wrong, the angle was wrong, or the whole place was now a parking lot, and drive home having learned nothing except that I need a better plan.

Now I do a chunk of that scouting from my computer. I'll describe what I'm after, something like a wide open field with a tree line for a sunrise shot within two hours of Huntsville, and get back a list of specific places worth checking, along with why they might work. Then I cross-reference that with satellite images and elevation maps to figure out which direction I'd actually be shooting and whether there's a hill blocking my sunrise before it clears the horizon. There have been many apps developed in recent years that help greatly with this process.

Weather

Landscape photography runs on a brutal schedule. You either get up before the sun or you don't get the shot, and there is no snooze button on golden hour. The problem is you don't find out if it was worth it until you're already standing there in the dark, getting cold, watching a flat gray sky do absolutely nothing.

I've started using AI tools to chew through weather data before I commit to the alarm clock. Cloud cover percentage alone doesn't tell you much. What I actually want to know is whether there's a break in the clouds low on the horizon that'll let the sun light up the underside of everything else, because that's the difference between a postcard and a gray soup.

I can feed in the forecast details and ask it to reason through what that combination is likely to produce, which is a lot faster than me squinting at five different weather apps trying to do the math myself at eleven at night.

Is it right every time? No. Nothing is right every time when you're talking about weather.

I still get skunked sometimes. I just get skunked less often, and I sleep in more mornings when the sky was never going to cooperate anyway.

This one's less romantic but it might save me the most time. Photography trips have a lot of moving parts that have nothing to do with photography. Sunrise and sunset times shift the whole day around. Some locations need a permit or have access windows. Driving between spots eats more daylight than you think it will, especially if you're hauling gear and stopping every time something looks interesting, which, if you're me, is constantly.

Trip Planning

Before a trip, I'll lay out what I'm trying to see and in what order, and use AI to help build out a realistic day by day plan, accounting for drive times, sunrise and sunset, and how much time I actually need at each spot instead of the fantasy version where everything takes fifteen minutes. It's basically like having someone build me a first draft itinerary that I then argue with and adjust, instead of starting from a blank page and a map with sixty tabs open.

I want to be clear about what this isn't. It's not a substitute for local knowledge, it's not always right, and it definitely doesn't replace showing up and reading the actual scene in front of you, because plans fall apart the second you get there, and the light is doing something the forecast didn't mention.

What it does is cut down on the wasted trips, the wasted alarms, and the wasted afternoons doing logistics instead of shooting.

I'd rather spend my planning time getting to good spots at the right time than spend my shooting time discovering I planned wrong. That's really all this comes down to. Less time guessing, more time standing in a field waiting for the light to do something worth photographing.

Shut up and show me what you are talking about

I use Claude for a lot of this, specifically a feature called Projects, which basically lets you build a standing set of instructions that stays in place every time you use it, instead of explaining yourself from scratch each time. I spent a while getting mine right, mostly through trial and error and a lot of "no, not like that." Here's a set of "Project Instructions" I have put together to get a lot of my scouting info quickly without searching across multiple apps every time.

Basically, I ask it to build me a scouting dashboard for one location and one date. It pulls sunrise, sunset, golden hour windows, and moon phase, then checks the weather for that day plus two days on either side, breaking cloud cover into low, mid, and high layers when that data exists, because a flat gray sky and a sky with high thin clouds are two completely different mornings even if the weather app just says "60% clouds" for both.

It also checks road and trail closures, permits, tripod bans, and drone rules, plus parking, admission, and whether the gate's even open before sunrise.

Then it lays all of that out as a set of visual cards with a status badge up top telling me if conditions are good, mixed, or a reschedule, and one line at the bottom telling me straight up whether to go.

List version:

• Give it a location and a date, nothing else

• It pulls live sunrise, sunset, golden hour timing, and moon phase

• Checks the weather for that day plus two days on either side, and breaks cloud cover into low, mid, and high layers when the data supports it, since "60% cloudy" can mean a washout or a sky on fire depending on which layer it is

• Flags road closures, permits, tripod bans, and drone restrictions

• Checks parking, admission, and whether the place is even open before sunrise

• Skips anything that doesn't apply instead of padding it out

• Says "I don't know" instead of guessing when the data isn't out there

• Wraps it all up as a scannable dashboard with a status badge and one line telling me whether the trip's worth it

Let me show you the output from a test run. I simply enter my desired location of Schwabacher's Landing and tell it to check what tomorrow morning, August 20, 2026, looks like for sunrise.

After about 5 minutes, this is what I am presented with:

So, should you actually try this? Yeah, probably. It won't scout the trail for you and it'll still get things wrong sometimes, especially in spots that don't have a lot of published data.

But for the boring parts, sunrise math, weather layers, permit rules, whether the parking lot fills up before you even wake up, it saves a genuine chunk of time and catches things I'd otherwise find out the hard way.

Worth a shot next time you're planning a shoot.