AI REGULATION: Bad AI poetry puts Biden in actors’ camp against fakes

It’s been one of the sticking points in the ongoing dispute between the Screen Actors Guild (SAG) and Hollywood and the question of copyright regulations and artificial intelligence has now become the focus of US President Joe Biden and the US Congress.

What’s worrying actors and legislators is the use of AI technology to create deepfakes that use the  images or voice of actors, singers and other performers without their consent. As SAG national executive director Duncan Crabtree-Ireland said in a statement this week, the “explosion in popularity and capability of generative AI has flooded the internet with songs, videos and voice recordings that exploit the voices and likenesses of our members without consent or compensation. For our members, their voice and likeness is their livelihood. They spend a lifetime improving their talent and building their value. It is outrageous to think someone can undermine that value with a few prompts and clicks on a keyboard.”

While a recent case in India saw actor Anil Kapoor successfully win a case that ensures that his “likeness, image voice or any other aspect of his persona” may not be used without his permission to create any digital content, in the US the question of copyright and AI has yet to be addressed on the legislative level.

That may soon change thanks to a new bipartisan bill by four senators this week. The “Nurture Originals, Foster Art and Keep Entertainment Safe” (NO FAKES) Act seeks to protect performers from unlicensed use of their image or likeness for commercial purposes and would allow companies or individuals who ignore it to be sued for violations. This is the first attempt by the US federal government to protect performers from the rapidly rising threats of AI to their livelihoods and under the proposed legislation the rights of both living and deceased artists would be protected. In the case of living performers rights would continue to exist 70 years after their deaths and in the case of already dead artists the rights would fall to their heirs or estates for the same period.

According to the Hollywood Reporter, the first draft of the NO FAKES Act specifically targets anyone who makes a “digital replica” without consent from the performer it uses and defines such a replica as “a newly created, computer-generated, electronic representation of the image, voice or visual likeness of an individual” that appears “nearly indistinguishable” from a performer’s actual characteristics and was embodied in a work in which that person never actually performed or appeared”.

The Motion Picture Association (MPA), which represents the interests of studios and content producers, is a little more wary of potential legislation than actors and other performers might want it to be. In a filing made with the US Copyright Office this week, the MPA laid out its positions on a number of AI-related issues and hedged its bets somewhat by arguing that “while AI technologies raise a host of novel questions, those questions implicate well-established copyright law doctrines and principles. At present, there is no reason to conclude that these existing doctrines and principles will be inadequate to provide courts and the Copyright Office with the tools they need to answer AI-related questions as and when they arise.”

The NO FAKES Act fits neatly into a broader recent legislative focus by the Biden government on AI and its threats that many are reporting was inspired, or at least given more urgent impetus, by the 80-year-old president’s recent watching of Tom Cruise’s latest Mission Impossible instalment Dead Reckoning in which hero Ethan Hunt and his team face off against a malevolent AI being, called “The Entity”, that attempts to take over global politics by manipulating digital information.

According to the Associated Press, Biden’s chief of staff Bruce Reed recalled that while the president had been increasingly curious and worried about the rise of AI and its implications, he was gripped by renewed urgency after he watched the film, and “If he hadn’t already been concerned about what could go wrong with AI before that movie, he saw plenty more to worry about”.  Biden was also shown examples of the abilities of AI, including “fake images of himself, his dog”, “bad poetry” and the “incredible and terrifying technology of voice cloning”, which helped to push him this week to issue an executive order that aims to place some control over the ways that tech companies are using AI technology.

Biden explained his order as an important measure towards putting in place policies that govern AI to “realise the promise and avoid the risks”.

You can be sure that this has come as a welcome sign to SAG members and perhaps it’s inspired other groups who have long attempted to get Biden to pay them much needed attention to get Cruise on board for films about climate change, gerrymandering, healthcare or the decline of the US Supreme Court. Of course they would have to pay Cruise to either act in these films or at the very least allow them to use his likeness for the purposes of campaigning for Biden’s attention.

The AI regulation baton has now passed to vice-president Kamala Harris, who this week travelled to the holy grail of British government technology, Bletchley Park, for the Global Summit on AI Safety where world leaders and tech executives including Elon Musk and Open AI founder Sam Altman were meeting to discuss draft policies on government and military use of AI and the setting up of a $200m philanthropic fund intended to “ensure that AI is used in the public interest”.

This may all seem simple enough but if the US Congress’s previous record on hearings involving tech is anything to go by — Facebook, Google, WhatsApp — you can probably bet that its public representatives will take the opportunity to woefully misinterpret or fail to understand yet another aspect of modern digital technology that’s popular with “the kids” and generate plenty of material for gifs and viral videos in the process.