X
1455 Rate this article:
5.0

Using LLMs To Research Remote Sensing Software: Helpful, but Incomplete

JP Metcalf

Whether you’re new to remote sensing or a seasoned expert, there is no doubt that large language models (LLMs) like OpenAI’s ChatGPT or Google’s Gemini can be incredibly useful in many aspects of research. From exploring the electromagnetic spectrum to creating object detection models using the latest deep learning methods, these tools can jumpstart your understanding and provide helpful context on a wide range of remote sensing topics.

LLMs seem to know a little about everything, and they’re fast. For instance, when I asked ChatGPT (4o model), “Compare and contrast the top 3 desktop remote sensing software products,” it returned a clean, nicely formatted table in seconds.

 

 

The table broke down features like user interfaces, spectral/SAR processing, machine learning capabilities, and expected learning curve. At first glance, it looked accurate enough. But as I dug in, it became clear: LLMs are great at sounding confident, but not always great at being complete. 

Having worked extensively with all three of the platforms listed, I noticed a few gaps and oversimplifications. For example, the Spectral Capabilities section mentions NDVI as a feature for ENVI® and ArcGIS Pro. While technically correct, that sells ENVI short. I’d rather highlight something with more depth, like ENVI’s newly designed Spectral Hourglass Workflow, which truly showcases its analytical power.

Similarly, the entry for photogrammetry described ENVI’s capabilities as “Basic.” Overlooking not only its dedicated module, but also Summit Evolution, a world-class photogrammetric workstation. On the bright side, the LLM did correctly mention our ImageIQ software, which offers powerful UAV-focused photogrammetry tools.

The SAR Tools section was another missed opportunity. It didn’t mention ENVI SARscape, ENVI SARscape Analytics, or the new ENVI SAR Essentials module, which covers everything from basic to advanced SAR analysis. 

One more row in the table for geospatial services and enterprise capabilities could’ve made a big difference. It would have introduced readers to the ENVI Ecosystem, which includes cloud-based solutions like ENVI Inform for automated processing and ENVI Connect for intuitive, web-based data visualization and analysis across an organization.

So no, the AI-generated table wasn’t perfect. But that doesn’t mean LLMs aren’t helpful. In fact, they shine in several key areas:

 

  • Getting a baseline when you’re unfamiliar with a topic.

  • Quickly generating side-by-side feature comparisons (like the table above).

  • Summarizing dense documentation into digestible highlights.

 

Still, LLMs won’t replace domain expertise. If you’re seriously evaluating software for your workflows, whether it’s flood mapping, hyperspectral mineral detection, or precision agriculture, you’ll need to go further. That means reading user guides, engaging in community forums, and most importantly, trying out the software yourself.

AI can give you a head start. But when it comes to making informed decisions about geospatial technology, there’s no substitute for hands-on experience.

 

If you’d like to discuss different remote sensing solutions, simply reach out to us.

 

From Image to Insight: How GEOINT Automation Is Changing the Speed of Decision-Making From Image to Insight: How GEOINT Automation Is Changing the Speed of Decision-Making Not All Supernovae Are Created Equal: Rethinking the Universe’s Measuring Tools Not All Supernovae Are Created Equal: Rethinking the Universe’s Measuring Tools