Post Thumbnail

How synchronization of 3 light sources protects against forgeries

Artificial intelligence has learned to create video fakes that are impossible to distinguish from reality. And this is a huge problem and question of trust in society. But scientists from Cornell University found a brilliant solution. They hid watermarks right in ordinary lighting.

The technology is called NCI. Which stands for noise-coded illumination. Professor Abe Davis and his team made light flicker in a special way. These tiny brightness fluctuations create a unique code that the human eye doesn’t notice. But the camera records every detail.

Imagine — ordinary lamps in the room or your computer screen transmit a secret code through barely noticeable brightness changes. And the light flickers according to a set pattern, creating an invisible watermark throughout the entire video.

And here’s how authenticity is verified. A computer with the key to the code analyzes the recorded video and generates its own version with timestamps from the light fluctuations. If the video is real — the versions match perfectly. If someone used a deepfake — obvious discrepancies appear. Black areas emerge in forged places or the image disappears completely.

But most interesting — different light sources can transmit different codes simultaneously. The chandelier broadcasts one code, the desk lamp another, the laptop screen a third. To fake such a video, a fraudster would have to recreate each light code separately. And they all must be perfectly synchronized with each other. This is practically impossible. Probably.

Unlike existing protection systems that require special cameras, NCI works with any device. From professional cameras to ordinary smartphones.

The technology is still in development. But one already wants them to succeed.

Autor: AIvengo
For 5 years I have been working with machine learning and artificial intelligence. And this field never ceases to amaze, inspire and interest me.
Latest News
XPeng introduced world's first female humanoid robot

Chinese electric car manufacturer XPeng introduced the new generation humanoid robot IRON. And this is the first female humanoid!

Michael Burry bet 1.1 billion dollars against Nvidia and Palantir

Michael Burry - this is a legendary investor who predicted the 2008 mortgage crisis. And now he's making a loud move again. Michael bet 1.1 billion dollars in put options against 2 major companies from the AI sector. These are Nvidia and Palantir.

Anthropic conducts interviews with models before sending to retirement

Anthropic published a policy for "decommissioning" outdated AI versions. Key commitment is to preserve weights of all public and actively used internal models for at least the company's lifetime. So that in the future access can be restored if necessary.

Nvidia head believes there is no AI bubble

Nvidia founder Jensen Huang dispelled concerns about a bubble in the AI market. And according to him, the company's latest chips are expected to bring 0.5 trillion dollars in revenue.

Sam Altman is tired of money questions

Sam Altman is tired of questions about OpenAI's money. And this became obvious during a joint interview with Satya Nadella on the Bg2 podcast.