AI-powered wildlife tracking and conservation

TEXT_INPUT:Ask the user to enter the location of interest;;HTTP_REQUEST:Fetch data about wildlife sightings in the specified location;;CHATGPT:Prompt AI to provide information about the species spotted most frequently;;DISPLAY_OUTPUT:Display the information about the most spotted species in a user-friendly format;;DALL_E:Generate an image of the most spotted species using DALL-E

Generative AI models such as Language Model GPT-3 (LLMs) and image generators can be harnessed significantly in wildlife tracking and conservation. For instance, LLMs can analyze vast amounts of text data from wildlife studies, tracking notes, and observations, providing insightful outputs that can guide conservation strategies by identifying patterns, habits, or changes in wildlife behaviours that might otherwise be too complex to glean manually.

Image generators working with Generative Adversarial Networks (GANs) can analyze thousands of trail camera images to recognize and classify different species. Furthermore, these AI models can be employed to predict wildlife movements based on past tracking data, facilitating efficient deployment of conservation resources.

Models like GPT-3 could also be trained to understand and detect specific calls or sounds from audio data to identify certain species. This can aid in real-time tracking and proactive measures to protect threatened species or habitats.

How to build with Clevis

This is an example of an application that you can build using Clevis, an AI application generation tool. The example given here aims at creating an AI-powered Wildlife Tracker. The intention of the application is to track and conserve wildlife by gathering data on wildlife sightings.

The first process-step Text Input is user-driven where the user is asked to enter their location of interest. Once the location information is entered, the second step, Http Request, fetches data about the sightings of wildlife in the specified location.

This data is then processed and analyzed in the ChatGPT step, which uses the OpenAI's ChatGPT model. In this phase, AI is prompted to provide information about the species spotted the most frequently. The findings derived from this stage are then displayed in a user-friendly format in the Display Output stage.

Finally, it engages DALL·E, another AI model by OpenAI, to generate an image of the most commonly spotted species. This creates a more intuitive and engaging UI.

Using Clevis for applications like this, allows one to automate and expedite the process of building applications within the wildlife tracking and conservation domain.