Technology
Friday, February 16th, 2024 6:13 pm EDT
Key Points
- A consortium of 20 leading tech companies has made a joint commitment to combat AI misinformation during the 2024 elections, focusing particularly on addressing the proliferation of deepfakes, which utilize deceptive audio, video, and images to manipulate democratic processes and disseminate false voting information.
- Signatories to the accord include major industry players such as Microsoft, Meta, Google, Amazon, IBM, Adobe, and chip designer Arm, as well as AI startups like OpenAI, Anthropic, and Stability AI, and social media platforms such as Snap, TikTok, and X.
- The announcement comes amid preparations for a significant year of elections worldwide, affecting over four billion people across more than 40 countries, where concerns about election-related misinformation have escalated due to the exponential increase in the creation of deepfakes, rising by 900% year over year. Despite this surge, detection and watermarking technologies for identifying deepfakes have not kept pace with advancements.
A consortium of 20 leading tech companies announced a collective commitment to combat AI misinformation during the 2024 elections, particularly targeting deepfakes, which utilize deceptive audio, video, and images to mimic influential figures in democratic processes or disseminate false voting information. Among the signatories to the accord are industry giants like Microsoft, Meta, Google, Amazon, IBM, Adobe, and chip designer Arm, alongside AI startups including OpenAI, Anthropic, and Stability AI, and social media platforms such as Snap, TikTok, and X. With numerous elections globally affecting billions of people across more than 40 countries in 2024, tech platforms are gearing up to tackle the escalating concerns surrounding AI-generated content, with deepfake creation witnessing a staggering 900% year-over-year increase, according to Clarity, a machine learning firm. Despite this, detection and watermarking technologies for identifying deepfakes have not kept pace with advancements. The announcement of the accord coincides with the unveiling of Sora, a new AI model for generating videos by OpenAI, designed similarly to its image-generation tool DALL-E. Sora enables users to describe desired scenes, prompting the generation of high-definition video clips or the extension of existing ones. The participating companies committed to eight key objectives, including assessing model risks, detecting and addressing the distribution of deceptive content on their platforms, and ensuring transparency in these processes. While these commitments are voluntary and applicable only where relevant for each company’s services, they underscore the industry’s dedication to safeguarding the integrity of elections. Kent Walker, Google’s president of global affairs, emphasized the crucial role of secure elections in maintaining democracy, while Christina Montgomery, IBM’s chief privacy and trust officer, stressed the need for concrete, cooperative measures to mitigate the amplified risks posed by AI-generated deceptive content in this pivotal election year.
For the full original article on CNBC, please click here: https://www.cnbc.com/2024/02/16/tech-and-ai-companies-sign-accord-to-combat-election-related-deepfakes.html
