⚡️ Introducing DETECT-2B, the most advanced deepfake detection model

Q

Advancing Ethical AI: Resemble AI’s Commitment to the AI Voluntary Code of Conduct

Dec 11, 2023

Resemble AI proudly announces its commitment to the Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems. This landmark initiative, launched by the Canadian government, aims to establish ethical guidelines and best practices for this rapidly evolving field. This step aligns Resemble AI with global efforts to ensure that AI benefits society while addressing risks, and it ties directly into our ongoing initiatives like the Neural Speech Watermarker and the Deepfake Detector, “Detect.”


The Voluntary Code of Conduct: A Step Forward

The Canadian Government’s initiative, led by the
Honourable François-Philippe Champagne, Minister of Innovation, Science and Industry, has attracted major AI organizations, including Resemble AI. We’re joined by IBM, Scale AI, CGI, Bluedot, and AltaML who have made the pledge as well. This code is a vital measure in fostering a responsible AI ecosystem in Canada and beyond, focusing on accountability, safety, fairness, transparency, human oversight, and robustness in AI systems.

Resemble AI’s Pioneering Initiatives: Neural Watermarker and Detect

Resemble AI’s dedication to ethical AI is further exemplified through our innovative products: the Neural Speech Watermarker and “Detect,” our deepfake detection tool. These initiatives are aligned with the code’s principles, emphasizing our commitment to secure, fair, and transparent AI systems.

The Neural Speech AI Watermarker

Our Neural Speech AI Watermarker is a state-of-the-art solution designed to secure AI-generated voice content. It embeds an inaudible watermark into audio files, ensuring the traceability and integrity of the content. This technology is a testament to our dedication to preserving intellectual property rights and preventing misuse of AI in creating deepfake voice content.

Detect: The Deepfake Detector

“Detect,” our advanced deepfake detection tool, represents a significant advancement in AI safety. It utilizes a cutting-edge neural network to differentiate between real and fake audio files, addressing the increasing concern of deepfakes in various sectors, from music to politics. Detect is designed to work in real-time, providing up to 98% accuracy in exposing deepfake audio, thereby safeguarding the authenticity of voice content.

Integrating Our Initiatives with the Code of Conduct

By joining the Voluntary Code of Conduct and integrating our initiatives like the Neural Watermarker and Detect, Resemble AI is not only adhering to but also leading the way in responsible AI development. These measures ensure that our technologies are safe, fair, and beneficial for society.

Accountability and Safety

In the realm of advanced generative AI systems, the twin pillars of accountability and safety are crucial. At Resemble AI, our commitment to these principles is not just a policy stance but a foundational aspect of our operational ethos, especially in light of our recent initiatives such as the Neural Speech Watermarker and the “Detect” deepfake detection tool.

Fairness and Equity

Fairness and equity are critical components in the development and deployment of AI systems, especially in the context of advanced generative AI like that of Resemble AI. These principles are not just ethical imperatives but also crucial for the practical effectiveness and societal acceptance of AI technologies.

Transparency and Human Oversight

Transparency and human oversight are indispensable in the realm of AI, particularly for companies like Resemble AI, which are at the forefront of generative voice AI technology. These concepts play a pivotal role in building trust, ensuring ethical usage, and facilitating understanding among users and stakeholders. The deployment of Detect and the Neural Watermarker epitomizes our commitment to transparency. These tools allow for greater insight into the authenticity of AI-generated content, ensuring informed decision-making and expert evaluation of risks.

Validity and Robustness

Our systems, bolstered by Detect and the Neural Watermarker, are designed to be robust and secure. They demonstrate our commitment to ensuring that AI operates as intended and remains secure against emerging threats.


Leading The Way In Ethical AI

Resemble AI’s participation in the Voluntary Code of Conduct, combined with our initiatives like Detect and the Neural Speech Watermarker, signifies our unyielding commitment to ethical AI development. As we continue to innovate in the voice AI space, we remain dedicated to principles that ensure our technologies are safe, fair, and beneficial for society. Our goal is to pave the way for a responsible and sustainable AI future, where innovation goes hand in hand with ethical responsibility.

More From This Category