In recent weeks, major tech companies have taken significant steps to address the growing problem of explicit deepfakes, potentially signaling a shift in the industry’s approach to this issue. This comes after sustained criticism from digital safety experts, women’s rights advocates, and legal scholars regarding Big Tech’s perceived role in the proliferation of non-consensual, sexually-explicit deepfakes online.

Meta’s Oversight Board ruled on two cases involving AI-generated explicit images of public figures, mandating their removal and recommending policy clarifications. The Board emphasized the severity of harms caused by such content and suggested replacing vague terms in content moderation policies with clearer language.

Microsoft published a report acknowledging the scale of the deepfake problem and calling for updated legislation to protect citizens from abusive AI-generated content. The company’s vice chair and president, Brad Smith, advocated for criminal penalties for creating and distributing non-consensual intimate images and child sexual abuse material generated by AI.

Google announced significant updates to its search engine to combat explicit deepfakes. These changes include overhauling the reporting process for victims and adjusting the search algorithm to push websites hosting deepfake pornography further down in search results. The company aims to surface high-quality, non-explicit content for queries specifically seeking this type of material.

While these developments have been welcomed by advocates fighting against image-based sexual abuse, it remains to be seen whether they will lead to better protections for victims and a reduction in the volume of such material online. However, these actions suggest that the tech giants are beginning to take the criticisms seriously and may be moving towards more comprehensive measures to address the deepfake abuse epidemic.

As the U.S. Congress considers legislation that would allow victims of “sexual digital forgeries” to sue their abusers, these initiatives from Big Tech may provide hope for a safer internet in the future.

Original Article

SEO Keywords: deepfakes, explicit content, Big Tech, content moderation, AI-generated images, digital safety, Meta, Microsoft, Google, online abuse

Google is implementing changes to its search engine to address the growing issue of deepfake pornography. The tech giant aims to make it easier for victims to remove non-consensual explicit content from search results. Key measures include omitting explicit results for searches containing a person’s name, downranking websites with high removal notice volumes, and automatically removing duplicates of removed images. Google reports a 70% reduction in explicit deepfake content in search results since early 2024. However, challenges remain, such as limited access to third-party media data and balancing the needs of adult performers. The move comes as tech companies face pressure to tackle the social impacts of generative AI, with regulators like the UK’s Online Safety Act making non-consensual deepfake pornography illegal. While advocacy groups push for complete delisting of deepfake sites, Google argues this could block access to important information about content removal.

Complete story here

SEO Keywords:

1. Deepfake pornography
2. Google search changes
3. AI-generated content
4. Online content removal
5. Non-consensual explicit material

This article discusses a recent deepfake video that falsely claimed Olena Zelenska, wife of Ukrainian President Volodymyr Zelensky, purchased a luxury Bugatti car during a visit to Paris. Key points include:

1. The video, which gained millions of views on social media, was identified as a likely part of a Russian-linked disinformation campaign.
2. Bugatti officially denied the claims and stated they were victims of criminal offenses, filing a legal complaint against the misinformation.
3. AI analysis indicated high certainty of facial manipulations in the video.
4. The video originated from a suspicious French-language website called “Verite Cachee France,” which appears to use AI-generated content.
5. Experts suggest this is part of a larger Russian narrative laundering campaign aimed at undermining Ukraine and eroding European support for the country.
6. The disinformation network uses a combination of AI-generated content, fake news websites, and social media amplification to spread pro-Russian propaganda.
7. This campaign is similar to previous efforts that spread false stories about Ukrainian officials, including a debunked claim about Olena Zelenska buying expensive jewelry.

The article highlights the growing concern over sophisticated disinformation campaigns using AI technology to create and spread false narratives for political purposes. 

Original Article Appears Here

At Euro 2024, England head coach Gareth Southgate’s statements are not only scrutinized but also fabricated using AI. Fake interviews with Southgate, created with AI voice generators, have gained millions of views on social media platforms like TikTok and Instagram. These videos falsely depict Southgate making derogatory remarks about players, including Phil Foden, Jordan Henderson, Jack Grealish, and Marcus Rashford. The Football Association has labeled these clips as “offensive” and is working to have them removed.

Some fans in Germany find the videos humorous, viewing them as harmless fun typical of British humor. However, concerns about the impact and ethics of AI-generated content persist. The FA emphasized that Southgate and the team remain focused on their upcoming match against Slovakia in the last 16, despite criticism of their performance.

Experts warn of the dangers posed by deepfake technology, not only for public figures but for anyone with a digital presence. Regulatory measures and societal consensus are needed to address the fine line between satire and harmful content.

For more details, you can read the full article on the BBC Sport website.

The article discusses the development of DIVID, a tool created by Columbia Engineering researchers to detect AI-generated videos. This innovation addresses the growing problem of highly realistic AI videos being used for scams. DIVID, short for DIffusion Video Detection, examines frames from diffusion-generated videos for inconsistencies that indicate AI manipulation. It builds on previous research involving Raidar, a tool for detecting AI-generated texts. The core method, DIRE (DIffusion Reconstruction Error), compares original frames to reconstructed ones to identify discrepancies, boasting a detection accuracy of up to 93.7%. This technology could potentially be integrated into platforms like Zoom to enhance real-time deepfake detection, offering a significant step forward in combating digital fraud and misinformation.

For further details, visit the full article here.