A.I. is reshaping content creation across social media and, therefore, livelihoods. Artificial intelligence is generating various content, from memes to video essays, and voiceovers of Reddit posts. Even news updates with millions of clicks such as “RealTalkingFish” on Instagram, a page with over 1.3 million followers, run by @Nicosmalls_33, uses A.I generated the voice of the realistic talking fish from the hit cartoon Spongebob Squarepants to distribute current events in short clips. Although A.I has its merits in advancing technology and scientific advancement, but it is a fair statement to say that it is a threat to creativity and the arts. This includes voice talent within the entertainment industry, as current real-life people are having their voices synthesized by strangers online, including some corporations.
Experts such as Hany Farid, a professor of digital forensics at the University of California at Berkeley, have long predicted that generative AI would lead to tons of faked images and videos, spreading fake news and making it harder to trust things that are seen online. NewsGuard identified 17 accounts on TikTok that made AI-generated videos using text-to-speech software that advanced the spread of advanced misinformation. These 17 accounts have achieved more than 336 million views and 14.5 million likes on the social media site. “You’re talking about violence, you’re talking about stealing of elections, you’re talking about fraud.” Says Farid. This has real-world consequences for individuals, societies, and democracies.”
It’s also important to mention that when voices are synthesized using audio of real-life people, it can be misleading to those who see the generated content. One can assume that it’s actually the person whose voice was sampled participated in a specific thing they have nothing to do with. One example was a YouTube channel that was creating AI-generated VR chat videos of characters from the indie animated show recently picked up from Netflix titled The Amazing Digital Circus. One of the voice actors on the show, Michael Kovach, who is a prominent voice actor in indie animation, made a tweet as a response when he found out about it. “Please do not AI synthesize my voice. I’ve been very lax about this up til now because it’s never sounded good,” he wrote. “But I just found a VRchat roleplaying video that genuinely sounds exactly like me with minimal artifacting. It actually made me feel genuine fear.”
Another prominent example of this that stuck out upon further research was this Youtube Channel with over 100,000 subscribers known as “Hell’s Hotspot”, formerly known as “Hazbin Hotspot”. It is a channel that produces A. I generated content based on the Amazon Prime Original known as Hazbin Hotel. The A.I generated content that included the synthesizing of voices of the characters within the animated series. Although animated, they’re voiced by real-life people. They write a disclaimer on the “About” section of their profile stating that they respect “the original voice actors, their boundaries, effort, and rights.” However, the voice actor of debatably the most popular character from the show the channel mimics, rightfully doesn’t appreciate content such as what “Hell’s Hotspot” makes.
Amir Talai, who voices the character named Alastor on the show, has made a handful of public statements against A. I generated content on X. “Ai is theft and stealing my voice even for ‘fun’ is not cute and I’m not cool with it.” He writes. “I recognize there’s not much I can do about it until governments step in to create fulsome regulation. But if you want to know how I feel, that’s it. Up to you whether you choose to respect it.”
He continues further in posts on the social media site, writing about how he’s “literally never seen a VA, actor, or visual artist be happy about their work being stolen and used for AI.” And even warned potential users of voice synthesizers about legal consequences. “If you are planning to steal someone’s work anyway, you may wanna do it from folks who are not connected to a major corporation with deep pockets and lots of lawyers”.
After Talai’s posts that are directly against the use of synthesized voices on X, the creator of Hell’s Hotspot changed the voice actor he would generate content with. They made an update video saying they were going to use the “pilot version” of the character, based on the 30-minute pilot that was released on YouTube back in 2019. On the same day of his announcement, Edward Bosco, the original voice of Alastor, wrote “I’m never okay with people using AI to replicate characters I play”, further proving Talai’s point that no actor or artist likes seeing their work being stolen by people using A.I to generate content for them.
Voice actors who voice iconic characters have also shown distaste for their voices being synthesized. Jim Cummings, the voice behind Winnie-The-Pooh and Looney Tunes’ Taz requested publicly for this voice to be removed from A.I voice generators. One of these companies known as UberDuck complied, and luckily accepted his complaints, removing his voice model from their system. Interestingly, UberDuck users posted on Reddit that they were upset about his removal.
Lawsuits related to A. I theft is being made as a recent string of legal action brought against various tech companies by artists, writers, and creatives overall. There are also concerns about intellectual property laws being violated and copyright infringements occurring since the work of others trains AI systems. A well-known recent example of this is the lawsuit against the AI Firm known as Lovo for allegedly creating clones of the voices of Linnea Sage and Paul Lehrman without their permission. “A tech company stole our voices, made clones of them, and sold them possibly hundreds of thousands of times.” says Linnea Sage. She was in the car with her husband, Paul, heard something strange while listening to a podcast about the Hollywood writers’ strike. “He (the host) was interviewing my voice. How disturbing and terrifying that moment was is hard to articulate”.
To conclude, A.I am undeniably reshaping the landscape of content creation, sparking severe ethical concerns about consent, originality, and ownership rights, including the right to one’s voice. Creators and talents of various professions feel their identities and work are being exploited without their permission. The call for stricter regulations and legal frameworks to be developed is perfectly reasonable to address. The line between innovation and exploitation will likely remain a contentious debate of ethics A.I and the future of art for decades to come within the 21st century.