Detecting deepfakes in real time: How Intel Labs uses AI to fight misinformation

A few years ago, deepfakes seemed like a new technology whose creators relied on serious computing power. Today, deepfakes are ubiquitous and have the potential to be misused for misinformation, hacking, and other nefarious purposes.

To combat this growing problem, Intel Labs has developed real-time deepfake detection technology. Ilke Demir, a senior research scientist at Intel, explains the technology behind deepfakes, Intel’s detection methods, and the ethical considerations involved in developing and implementing such tools.

Too: AI Ethicist Says Today’s AI Boom Will Exacerbate Social Problems If We Don’t Act Now

Deepfakes are videos, speeches or images where the actor or action is not real but created by artificial intelligence (AI). Deepfakes use complex deep-learning architectures, such as generative adversarial networks, variable auto-encoders, and other AI models, to create highly realistic and believable content. These models can generate synthetic personas, lip-sync videos and even text-to-image conversion, making it challenging to distinguish between genuine and fake content.

The term deepfake is sometimes applied to authentic material. changelike this 2019 video from former House Speaker Nancy Pelosi, who was twisted to make him appear intoxicated.

Demir’s team investigates computational deepfakes, which are synthetic forms of material generated by machines. “The reason it’s called a deepfake is because generic AI has this complex deep-learning architecture that produces all that stuff,” he says.

Too: Most Americans think AI is a threat to humanity, according to a poll

Cyber ​​criminals and other bad actors often misuse deepfake technology. Some use cases include political misinformation, adult content featuring celebrities or non-consenting individuals, market manipulation, and impersonation for monetary gain. These negative effects emphasize the need for effective deepfake detection methods.

Intel Labs has developed one of the world’s first real-time deepfake detection platforms. Instead of looking for artifacts of counterfeiting, the technology focuses on detecting what’s real, such as heart rate. Using a technique called photoplethysmography – the detection system analyzes color changes in veins due to oxygen content, which are visible computationally – the technology can detect whether a personality is a real human or a synthetic one.

Demir said, “We are trying to see what is real and authentic. Heart rate is one of the (signs).” “So when your heart pumps blood, it goes into your veins, and the amount of oxygen causes the veins to change color, causing that color change. It’s not visible to our eyes; I just can’t see this video.” Can’t see nor see your heart rate. But that color change is visible computationally.”

Too: Don’t fall for fake ChatGPT apps: Here’s what to watch out for

Intel’s deepfake detection technology is being implemented across a variety of sectors and platforms, including social media tools, news agencies, broadcasters, content creation tools, startups and non-profits. By integrating the technology into their workflow, these organizations can better identify and reduce the spread of deepfakes and misinformation.

Despite the potential for abuse, there are legitimate applications of deepfake technology. One of the earliest uses was the creation of avatars to better represent individuals in digital environments. Demir refers to a specific use case called “MyFace, MyChoice”, which leverages deepfakes to enhance privacy on online platforms.

In simple terms, the approach allows individuals to control their appearance in online photos by replacing their faces with “quantitatively different deepfakes” if they wish to avoid being identified. These controls provide increased privacy and control over one’s identity, helping to counteract automatic face-recognition algorithms.

Too: GPT-3.5 vs GPT-4: Is ChatGPT Plus Worth Its Subscription Fee?

It is important to ensure the ethical development and implementation of AI technologies. Intel’s trusted media team collaborates with anthropologists, social scientists and user researchers to evaluate and refine the technology. The company also has a Responsible AI Council, which reviews AI systems for responsible and ethical principles, including potential biases, limitations, and potentially harmful use cases. This multidisciplinary approach helps ensure that AI technologies, such as deepfake detection, benefit humans rather than harm them.

Dimmer says, “We have legal people, we have social scientists, we have psychologists, and they’re all coming together to figure out whether there’s a bias — algorithmic bias, systematic bias, data Prejudice, any kind of prejudice.” , The team scans the code to find “any possible use cases of technology that could harm people”.

Too: 5 Ways to Explore the Use of Generative AI at Workplace

As deepfakes become more prevalent and sophisticated, the development and implementation of detection techniques to combat misinformation and other harmful consequences is becoming increasingly important. Intel Labs’ real-time deepfake detection technology provides a scalable and effective solution to this growing problem.

By incorporating ethical considerations and collaborating with experts across disciplines, Intel is working towards a future where AI technologies are used responsibly and for the good of society.