In a new development, OpenAI, known for creating ChatGPT, introduced a voice cloning tool named “Voice Engine” on Friday.
With a commitment to security, the company will keep this technology under strict control.
The tool mimics speech using just a 15-second voice sample, aiming to prevent the creation of misleading audio content.
The San Francisco-based entity emphasized the potential risks, especially concerning disinformation in critical times like election years.
It seeks broad feedback from sectors worldwide to refine the tool responsibly.
Experts in disinformation highlight the challenge posed by advanced, hard-to-detect AI voice cloning technologies.
Such advancements come at a crucial time when information authenticity is crucial. OpenAI’s cautious rollout responds to increasing concerns.
For instance, a misleading robocall imitating President Biden influenced the New Hampshire primaries. This event underscored the potential for AI to fuel misinformation campaigns.
To counter misuse, OpenAI has partnered with entities that agree to enforce rigorous standards, including obtaining explicit consent from individuals whose voices are cloned.
Additionally, the company is committed to maintaining transparency regarding using AI-generated voices.
Security measures, such as watermarking and usage monitoring, are in place to ensure the tool’s safe application.
OpenAI’s initiative reflects a thoughtful innovation approach while addressing ethical and safety concerns.