年轻人迎接挑战，用真相对抗深度造假威胁

世界经济论坛《2024 年全球风险报告》发出了严厉警告，指出主要由深度造假造成的错误信息和虚假信息是未来两年世界面临的最严重的全球短期风险。

与此同时，魁北克创新委员会在与专家和公众进行数月磋商后也认识到了局势的严重性。

打击 Deepfake 的挑战

Deepfakes 利用人工智能和生成式人工智能来创造超现实的制造品，其超越了技术奇迹，构成了深远的社会威胁。

尽管深度造假检测工具和立法工作不断发展，但在有效打击这一威胁方面仍然存在巨大差距。

虽然 Deepfake 检测工具正在开发中，但它们很难跟上 Deepfake 算法快速发展的能力。

法律体系和政府也在努力在打击数字欺骗的斗争中保持领先地位。

鉴于这些挑战，显然仅靠技术和立法不足以应对这一迫在眉睫的威胁。

政治虚假信息担忧

深度造假最令人担忧的方面之一是它们有可能加剧政治两极分化。

预计未来两年内将有近 30 亿人在各国投票，虚假信息运动可能会破坏新当选政府的合法性。

巴勒斯坦美国超模贝拉·哈迪德等知名人物的深度造假被操纵来伪造他们的政治声明，展示了该技术影响公众舆论和扭曲政治叙事的能力。

Greta Thunberg 倡导“纯素手榴弹”的 Deepfake 就是典型的 Deepfake 恶意使用的例子。

金融诈骗和色情危害

除了政治操纵之外，虚假视频还被用来进行金融欺诈。

值得注意的是，YouTube 知名博主 MrBeast 在 TikTok 上被冒充深度造假诈骗，导致财务欺诈。

此外，深度伪造色情内容引起了严重关注，因为个人的面孔未经同意就叠加在露骨内容上。

泰勒·斯威夫特 (Taylor Swift) 的色情深度伪造图片在社交媒体上传播等实例凸显了人工智能驱动的针对所有年龄段个人的欺诈和诈骗的脆弱性。

政策和技术方法

各国和科技巨头都出台了打击假新闻的政策和技术。

Meta 和谷歌等公司已要求政治广告商披露其广告中的人工智能操纵情况。

Additionally, researchers are developing advanced detection tools for deepfake audio, and legislative acts, such as the Deepfakes Accountability Act, have been introduced in the U.S. and Canada. The European Union has also taken steps to address online safety and AI regulation through the AI Liability Directive.

Research study with youth

To effectively combat the rising tide of disinformation fueled by deepfakes and AI technologies, a research project funded by the Social Sciences and Humanities Research Council (SSHRC) and Canadian Heritage focused on empowering youth with digital agency.

The study included Canadian university students aged 18 to 24, who participated in workshops, interviews, and focus group discussions.

Through hands-on experiences creating deepfakes, participants gained valuable insights into the technology and its potential for misuse. Initially perceiving deepfakes as uncontrollable, they progressed to understand their grave threat and developed a sense of responsibility in preventing their spread.

Youth empowerment recommendations

In the ongoing battle against deepfake misinformation, insights from recent research offer valuable recommendations for effective action:

Education plays a pivotal role by integrating deep fake awareness into curricula. Teaching the consequences of disinformation equips students to critically assess content.

Encouraging critical thinking is vital. Providing spaces for youth to question societal norms and educating them about social media policies fosters responsible online behavior.

Technology awareness is crucial. Teaching students to recognize deepfakes by unveiling the creation process enhances their ability to spot deceptive content.

Active involvement is encouraged. Youth should engage in meaningful causes while remaining vigilant against disinformation, and they should be guided in addressing false information constructively.

A multifaceted strategy

To counter the proliferation of deepfakes effectively, a multifaceted strategy is proposed. This includes integrating deepfake education into curricula, nurturing critical thinking and digital agency in youth, and encouraging active yet well-informed participation in the fight against malicious deepfakes.

An interdisciplinary educational approach, combining technology, psychology, media studies, and ethics, is also emphasized to fully comprehend the implications of deepfakes.

The human element in combating Deepfakes

This research underscores the vital role of education and the human element in the fight against deepfakes. Relying solely on technology and legal solutions is insufficient. By equipping younger generations and society at large with the skills to critically analyze and challenge disinformation, we can nurture a digitally literate and resilient society capable of withstanding the manipulative power of fake news.

Ultimately, educating individuals about their roles and agency in safeguarding the integrity of our digital world is essential.