It’s a timely decision. As generative AI becomes more integrated into filmmaking, debates over creativity and authorship are intensifying. Writers’ strikes and fears of artistic displacement have dominated recent industry discussions.

But how do audiences feel about the use of AI in films? Our research suggests they may be more open to it than the industry might expect.

What the new rules say

The updated Oscars guidelines make it clear the use of generative AI will neither help, nor hinder, a film’s chances of nomination.

What matters is the degree to which people remain at the centre of the creative process. While AI tools can be part of the workflow, the judges will scrutinise the standard of human creative authorship in a given work.

This reflects broader shifts taking place in the film industry. AI tools are now embedded in many stages of production, including for high-profile and award-nominated films.

At this year’s Oscars, Adrien Brody won best actor for his performance in The Brutalist, which used generative AI to enhance the actor’s Hungarian dialogue. Emilia Pérez – the most nominated film, with 13 nods – also used AI-powered voice cloning in post-production.

The Oscars update isn’t introducing AI to Hollywood. It’s simply acknowledging the extent to which it is already in use.

Do audiences mind?

To understand how audiences respond to AI’s creative role in film, we conducted an experiment testing people’s reactions to AI-generated film ideas.

For our study, published in the Journal of Cultural Economics, we asked 500 US-based participants to rate AI-generated film “pitches” in terms of their anticipated enjoyment and likelihood of watching the film across different formats (such as cinema, online rental, or streaming).

Half of the participants were explicitly told the ideas were generated by AI, while the other half were not. Each AI-generated pitch included a synopsis, director, top-billed cast, genre, rating and runtime.

The results were clear. There was no systematic bias against AI-generated pitches. Ratings of anticipated enjoyment and likelihood of watching the films were broadly similar, regardless of whether the participants knew AI was involved.

AI-assisted versus AI-produced

It’s important to note our research focused on audience reactions to ideas – the initial pitch for a film – and not the final product. This distinction matters.

AI’s role was limited in our experiment. Human directors and cast members were implicitly part of each pitch, and there was no suggestion AI had written the full screenplay or contributed in other ways to the production of the final film.

As we note in our paper, AI’s limited involvement likely shaped participants’ responses. There was an implicit understanding that human creativity would remain central to the final product.

This aligns with broader evidence from other creative sectors. In the case of music and visual art, audiences tend to respond less favourably when they believe a work has been fully AI-generated.

Together, these findings suggest the middle ground may be the best approach. While audiences may be accepting of AI’s contribution to creative tasks such as idea generation, editing, and visual and audio effects, they still value human authorship and authenticity in the final product.

That is also the balance the Academy Awards seems to be aiming for. The new rules do not disqualify films for using AI. However, they emphasise that awards will go to works where humans remain at the heart of the creative process. For now, audiences appear to be comfortable with that approach, too.

What it means for the industry

Generative tools are becoming part of the mainstream production toolkit. And this raises important questions about creative labour, credit and compensation.

While our research suggests audiences may be open to AI-generated content, this doesn’t mean the industry can move forward without careful deliberation. The question is no longer whether AI will shape the future of film, but how – and who gets to decide the terms.

Read More: AI voice technology used in The Brutalist is nothing new – the backlash is about transparency

If AI is to complement, rather than diminish, the filmmaking process, it will be important to maintain clear standards and ethical guidelines around AI use, as well as a clear role for human authorship.

This includes transparency around how AI tools are used, and appropriate recognition for creative contributions – including for those whose work has been used to train generative AI systems.

The real test will be whether the industry can embrace AI without losing sight of the creative values that define it.

• Paul Crosby is a Senior Lecturer, Department of Economics, Macquarie University
• Jordi McKenzie is an Associate Professor, Department of Economics, Macquarie University
• This article first appeared in The Conversation

The post The Oscars have rolled out the red carpet for generative AI. And surprisingly, viewers don’t seem to mind appeared first on Stuff South Africa.

These workers are the lifeblood of the burgeoning artificial intelligence (AI) industry. Without them, products such as ChatGPT simply would not exist. That’s because the data they label helps AI systems “learn”.

But despite the vital contribution this workforce makes to an industry which is expected to be worth US$407 billion by 2027, the people who comprise it are largely invisible and frequently exploited. Earlier this year nearly 100 data labellers and AI workers from Kenya who do work for companies like Facebook, Scale AI and OpenAI published an open letter to United States President Joe Biden in which they said:

Our working conditions amount to modern day slavery.

To ensure AI supply chains are ethical, industry and governments must urgently address this problem. But the key question is: how?

What is data labelling?

Data labelling is the process of annotating raw data — such as images, video or text — so that AI systems can recognise patterns and make predictions.

Self-driving cars, for example, rely on labelled video footage to distinguish pedestrians from road signs. Large language models such as ChatGPT rely on labelled text to understand human language.

These labelled datasets are the lifeblood of AI models. Without them, AI systems would be unable to function effectively.

Tech giants like Meta, Google, OpenAI and Microsoft outsource much of this work to data labelling factories in countries such as the Philippines, Kenya, India, Pakistan, Venezuela and Colombia.

China is also becoming another global hub for data labelling.

Outsourcing companies that facilitate this work include Scale AI, iMerit, and Samasource. These are very large companies in their own right. For example, Scale AI, which is headquartered in California, is now worth US$14 billion.

Cutting corners

Major tech firms like Alphabet (the parent company of Google), Amazon, Microsoft, Nvidia and Meta have poured billions into AI infrastructure, from computational power and data storage to emerging computational technologies.

Large-scale AI models can cost tens of millions of dollars to train. Once deployed, maintaining these models requires continuous investment in data labelling, refinement and real-world testing.

But while AI investment is significant, revenues have not always met expectations. Many industries continue to view AI projects as experimental with unclear profitability paths.

In response, many companies are cutting costs which affect those at the very bottom of the AI supply chain who are often highly vulnerable: data labellers.

Low wages, dangerous working conditions

One way companies involved in the AI supply chain try to reduce costs is by employing large numbers of data labellers in countries in the Global South such as the Philippines, Venezuela, Kenya and India. Workers in these countries face stagnating or shrinking wages.

For example, the hourly rate for AI data labellers in Venezuela ranges from between 90 cents and US$2. In comparison, in the United States, this rate is between US$10 to US$25 per hour.

In the Philippines, workers labelling data for multi-billion dollar companies such as Scale AI often earn far below the minimum wage.

Some labelling providers even resort to child labour for labelling purposes.

However, there are many other labour issues within the AI supply chain.

Many data labellers work in overcrowded and dusty environments which poses a serious risk to their health. They also often work as independent contractors, lacking access to protections such as health care or compensation.

The mental toll of data labelling work is also significant, with repetitive tasks, strict deadlines and rigid quality controls. Data labellers are also sometimes asked to read and label hate speech or other abusive language or material, which has been proven to have negative psychological effects.

Errors can lead to pay cuts or job losses. However, labellers often experience a lack of transparency on how their work is evaluated. They are often denied access to performance data, hindering their ability to improve or contest decisions.

AI workers in Kenya are organizing on multiple fronts, calling attention to how U.S. tech companies like ScaleAI & Meta are undermining labor laws.

Read their letter calling on President Biden to fight for improved pay, conditions, and safety on the job: https://t.co/2YENUOzjZk

— AI Now Institute (@AINowInstitute) May 23, 2024

Making AI supply chains ethical

As AI development becomes more complex and companies strive to maximise profits, the need for ethical AI supply chains is urgent.

One way companies can help ensure this is by applying a human right-centred design, deliberation and oversight approach to the entire AI supply chain. They must adopt fair wage policies, ensuring data labellers receive living wages that reflect the value of their contributions.

By embedding human rights into the supply chain, AI companies can foster a more ethical, sustainable industry, ensuring that both workers’ rights and corporate responsibility align with long-term success.

Governments should also create new regulation which mandates these practices, encouraging fairness and transparency. This includes transparency in performance evaluation and personal data processing, allowing workers to understand how they are assessed and to contest any inaccuracies.

Clear payment systems and recourse mechanisms will ensure workers are treated fairly. Instead of busting unions, as Scale AI did in Kenya in 2024, companies should also support the formation of digital labour unions or cooperatives. This will give workers a voice to advocate for better working conditions.

As users of AI products, we all can advocate for ethical practices by supporting companies that are transparent about their AI supply chains and commit to fair treatment of workers. Just as we reward green and fair trade producers of physical goods, we can push for change by choosing digital services or apps on our smartphones that adhere to human rights standards, promoting ethical brands through social media, and voting with our dollars for accountability from tech giants on a daily basis.

By making informed choices, we all can contribute to more ethical practices across the AI industry.

• Ganna Pogrebna is an Executive Director, AI and Cyber Futures Institute, Charles Sturt University
• This article first appeared in The Conversation

The post AI is a multi-billion dollar industry. It’s underpinned by an invisible and exploited workforce appeared first on Stuff South Africa.