OpenAI to Introduce Tool to Detect DALL-E 3 Generated Images

OpenAI plans to introduce tamper-resistant watermarking to mark digital content such as photos or audio, making it difficult to remove.

OpenAI announced the launch of a tool designed to detect images generated by its text-to-image generator DALL-E 3, a startup backed by Microsoft, on Tuesday, amidst growing concerns about the impact of AI-generated content on global elections this year.

According to the company, the tool demonstrated an accuracy of about 98% in identifying images created by DALL-E 3 during internal testing.

OpenAI Logo (Credits: OpenAI)

It has the capability to handle common modifications like compression, cropping, and saturation changes with minimal impact.

In line with these initiatives, OpenAI has joined an industry group that includes major tech players like Google, Microsoft, and Adobe. Together, they aim to establish a standard for tracing the origin of different media.

To support AI education and address societal challenges related to AI, OpenAI announced its collaboration with Microsoft in launching a $2 million “societal resilience” fund.

The move comes in the wake of fake videos circulating online during the recent general election in India, where two Bollywood actors were portrayed criticizing Prime Minister Narendra Modi.

The proliferation of AI-generated content and deepfakes has become increasingly prevalent not only in India but also in elections worldwide, including those in the U.S., Pakistan, and Indonesia.

Olivia Murphy
Driven by a commitment to excellence and integrity, Olivia strives to empower her audience with knowledge that enables informed decision-making and fosters a deeper understanding of the business world. She believes in the power of storytelling to bridge gaps, spark dialogue, and drive meaningful progress within the global business community.