OpenAI's Codex Interfaces with Adobe Lightroom to Denoise 50 Photos Without API

OpenAI's Codex, an AI coding service, interfaced with Adobe Lightroom to perform batch denoising on 50 photos. ai, reported using Codex to automate this task. The process avoided manual editing of each image individually.

Gostev described the task as requiring clicks into each photo for denoising, which he assigned to Codex. Codex determined the method to batch process the images. Business Insider reported that the service completed the task successfully.

Codex achieved this without using an official API, plugin, or browser-based approach. Instead, it interfaced directly with the Lightroom desktop application. No specific details on the interfacing method were provided in the report.

Task Denoising in Adobe Lightroom typically involves processing images one at a time.

This can be time-consuming for larger sets, such as the 50 photos Gostev handled. Automating such tasks through AI coding services represents an application of AI in photo editing workflows. Gostev, who is technically proficient, noted that he did not initially know how to batch the process manually.

He relied on Codex to identify and execute the steps. The incident occurred while Gostev was editing photos in Lightroom.

This example demonstrates Codex navigating and operating desktop software autonomously.

The service performed the task faster than manual methods would allow. Future developments in AI agents may expand such direct software interactions. ai focuses on AI capabilities, and Gostev's role involves evaluating AI performance.

The report originates from an interview with Gostev conducted by Business Insider. No further details on the exact timeline or replication steps were included.