Many serious photographers have got their pitchforks out regarding AI because of the vast arguments about how AI models are trained; at worst, it is copyright theft on a mass scale and at best... well let's save that discussion for another day.
Having read some of the discussion on how photographers would be losing out thanks to AI image generation, it got me thinking about how photographers could turn the tables and benefit from AI, without image creation or editing being part of the equation.
Wanting to get the best possible information, I sat down and talked with an expert, Claude.ai.
‘Besides image generation, manipulation and editing, what are some of the ways that AI can benefit photographers?’
I already knew the kind of answers I could expect and so wasn’t surprised when ‘Write model releases and other documents’ and ‘Write social media captions for images’ were amongst the replies. However, one suggestion caught my attention enough to investigate it further.
‘Create detailed location scouting reports combining weather, lighting, and timing data’
This was something so obvious, but that I had never considered. I had seen numerous articles about how people were asking AI to suggest their next holiday location based on their demands, and even to then make an itinerary for them, but it had never occurred to me that the same could be done for photography.
DIshing out advice
Intrigued I asked,
‘What considerations would you make if you were a photographer wanting to visit a location to take photographs at the best possible time? I want you to think about the needs of a landscape or travel photographer?’ (The irony isn’t lost on me that AI could have phrased that question a lot better).
The list of considerations that I got back was comprehensive; it had seemingly thought of everything. I’ll spare you the written response and put it as an image, otherwise I will go way over the word count for this article. However, it covered everything you expect such as sun position, sunrise and sunset times, typical weather etc, but impressively included local weather phenomena, migration patterns of wildlife, flowering/ foliage season for plants, tide schedules, tourist peak seasons, religious and cultural ceremonies, road conditions, avalanche risks… the list went on and on.
So I thought I would test its knowledge by asking what the best time of year is to visit somewhere on most photographer's bucket list of destinations – Jökulsarlon in Iceland. Having visited there myself, I have some knowledge of the location and was interested to see it’s response.
‘Late August, specifically around 4:30-5:30am’
That wasn’t the end, of course Claude.ai explained its findings with 26 different reasons. Leaving aside the weather considerations, among the most interesting reasons were that tourist crowds would be lower, roads would be fully accessible, there would be the potential for sunrise and aurora within the same shooting session, birds and seals would be active in the area, and August has good ice flow from the glacier.
However, I wasn’t satisfied. I wanted an exact date and time for the year 2025.
‘I calculate that August 23rd, 2025 at 4:45 AM as the optimal time, and here’s why:’
Again, there was a huge list of explanations, including all those mentioned above. It even mentioned that the low tide in the morning was better to photograph the ice formations on the nearby Diamond Beach. It also mentioned that the ‘Pre-weekend timing means fewer early-morning photographers’, but this wasn’t quite right.
AI isn’t infallible. As a language model it is basically looking at words and making a calculation as to what word comes next. It transpires that whilst the information was all generally correct, and logically sound, some of the specifcs were off. First off, August 23rd next year is a Saturday, so it can’t be ‘pre-weekend’. It also said that sunrise would be at 5:39 AM, which is close, but is actually 5:21 AM. Also, it suggested that the moon phase would be in the first quarter, but that day next year is actually a new moon, so whilst technically it is in the first quarter, it will be barely visible, which will actually make it great for astrophotography. The average temperature it suggested as 9°C (48°F) was accurate, as was the suggestion the chance of rain would be less than in September. The suggestion that the tide would be low was also reasonable, given the exact low point will be 7:57.
All of the more logical stuff was well thought out; it would still be in the summer travel season so easy to get to Iceland and get around, however, the season would be dying down, and the lagoon a lot quieter at that time of the morning before the tour buses start to arrive at around 8am. This also means the car park will quiet. It also said that it would allow photographers to capture the first light on the icebergs and capture multiple compositions as the light changes during sunrise.
Claude AI vs ChatGPT
For balance, I also did the same experiment with ChatGPT. It gave me a different date,
‘A great date and time to visit Jökulsárlón in 2025 would be Thursday, March 27, 2025.’
It gave me less reasoning to do with logistics and concentrated purely on photography and a schedule it suggests ‘…favorable conditions with a balanced partial moon, optimal aurora visibility, and manageable weather for winter photography.’
Back over at Claude AI, I corrected it on its mistakes and it revised the suggested date to 4:00 AM on Thursday 21st 2025. I then asked it for an itinerary, which you can see at the end of this article.
Although the information I was given wasn't perfect, it was close enough, and more importantly, so was the rationale behind its 'thinking'. We might not be quite there yet in terms of being able to rely 100% on AI, but as AI gains more access to API's and relevant websites, we are probably less than a year away from being able to simply ask the best time to take photographs and have accurate live data being pulled in. Especially as you can train AI to know your preferences for weather, lighting and locations.
I've already begun using AI to help me plan the best time of year to take certain photos here in the UK, and whilst I have come across a few inaccuracies, it has also opened my eyes to things that I haven't even considered when visiting a location.
Planning a photography trip will never be the same again.
The Itinerary:
Here's a detailed itinerary for photographing Jökulsárlón on Thursday, August 21st, 2025:
4:00 AM - Departure
Leave from nearby accommodation (recommend staying in Höfn or closer)
Weather and aurora check before departure
Gear check including extra batteries (morning temps around 8°C/46°F)
4:30 AM - Arrival & Initial Setup
Park in main lot while dark
Quick scout of lagoon conditions and ice positions
Setup primary camera for dark sky/aurora if conditions permit
Check tide times for Diamond Beach access
4:30-4:45 AM - Dark Sky Photography
Focus on any aurora or star reflections in lagoon
Use higher ISOs while dark (likely 3200-6400)
Consider light painting some foreground icebergs
4:45-5:00 AM - Pre-Dawn Blue Hour
Shift to lagoon compositions
Begin longer exposures of water movement
Look for ice formations catching first light
Consider panoramic sequences
5:00-5:21 AM - Civil Twilight
Move to pre-scouted primary composition
Capture ice colors emerging
Watch for wildlife (seals, birds)
Consider grad ND filters as light increases
5:21-5:45 AM - Sunrise
Capture direct light on ice
Move to secondary compositions
Focus on backlighting through ice
Watch for light beams through ice structures
5:45-6:15 AM - Early Morning Light
Quick move to Diamond Beach
Capture ice on black sand with early light
Work with wave movements around ice
Consider both wide and telephoto compositions
6:15-7:00 AM - Golden Hour
Return to lagoon for different light angle
Capture reflections in calmer water
Detail shots of ice textures in good light
Final wider compositions with established light
7:00 AM - Wrap Up
Final shots before tourist arrival
Consider staying for different light if conditions exceptional
Head to breakfast while reviewing images
Key gear considerations for this timeline:
Wide angle (16-35mm equivalent) for dark sky
Mid-range zoom (24-70mm) for main compositions
Telephoto (70-200mm) for ice details
Sturdy tripod for long exposures
ND and polarizing filters
Waterproof boots for beach work
Headlamp with red light option
Weather protection for gear
You might also like
