Uncategorized

• India is hurtling towards an election this year, and AI is playing a key role
• Deep fake media has become commonplace and its impacts escalating
• People have been arrested for forging videos of their political rivals

India’s 2024 general election, the largest democratic exercise in the world with over 960 million eligible voters, is being transformed by the rise of AI intelligence and hyper-realistic deep fake content. 

With this, a new generation of tech-savvy political operatives is leveraging AI to create synthetic media with political and commercial intent. 

One of the most prominent figures in India’s emerging ‘deep fake industry’ is Divyendra Singh Jadoun, a 31-year-old self-styled “Indian Deep faker” who has quickly become a go-to provider for political campaigns.

What started as a solo operation making Bollywood spoofs has evolved into a full-fledged synthetic media studio. 

“He’s since expanded because of all of the work that is coming to him,” Bangalore-based reporter Saritha Rai tells Bloomberg. 

“The first time I spoke to him, he said he had a couple of workers, and that was a few months ago. Then, he told me he has five other employees who help them make these videos.”

Using AI models trained on hours of authentic footage, Jadoun and his team generate uncannily realistic videos of politicians like Prime Minister Narendra Modi appearing to address voters by name. 

“The level of hyper-personalization that is possible with generative AI…You can just train the model to say the main message and then feed it with thousands and thousands of names. And it will instantly reproduce that same video and sync it perfectly with Narendra Modi calling out each person by name,” explains Rai.

AI election campaigns mimic the intimacy of in-person interactions at an almost infinite scale and affordable cost. 

Rai further notes, “a lot of the campaigning is going to be hyper personalized to the electors…a lot of it is going to reach them via their phones. And a lot of it will be AI generated material.”

Deep fakes are rife in India ahead of the election

Compounding the issue in India, Rai explains how millions of people are introduced to technology for the first time through affordable smartphones. 

A digitally inexperienced population is far more susceptible to deceptive content.

The same can also be said of older people, who, again, aren’t so exposed to the foolery of internet content.

But even among those who rate themselves as effective at identifying AI fakes, studies reveal that even when we know media is fake, it can affect our memory and decision-making. 

In addition to deep fakes designed to paint public figures in a positive light, instances of malicious political deep fakes are also becoming more common in India. 

For example, in November, fact-checker Muralikrishnan Chinnadurai spotted a livestream purporting to show Duwaraka, the long-dead daughter of a Tamil militant leader, giving a fiery political speech. 

Chinnadurai identified glitches, revealing it to be an AI-generated avatar. “This is an emotive issue in the state [Tamil Nadu] and with elections around the corner, the misinformation could quickly spread,” he told the BBC.

Last month, deep fake videos of Bollywood stars Ranveer Singh and Aamir Khan seemingly endorsing the opposition Congress party went viral, prompting police complaints from the actors. 

Credit: https://dailyai.com/2024/05/ai-deep-fakes-ignite-fierce-debate-ahead-of-the-2024-india-election/

Some scholars and media practitioners in Africa have called for the incorporation of training in artificial intelligence (AI) into the curriculum from elementary or primary to tertiary education levels, to raise a generation of African students grounded in developing AI systems that can enhance socio-economic development on the continent.

They said there is a need to train young Africans to become machine-learning experts and AI application developers, who will give primacy to African languages, cultures and traditions in the deployment of technologies to solve challenges confronting the continent.

In recent times, generative AI has caused huge disruptions in the way and manner of doing things in virtually every aspect of the human endeavour, creating remarkable advancements in diverse domains, including the creative sector, education and medicine.

However, Africa’s people are largely seen as consumers rather than active players in the scheme of things, a trend several participants across the continent recommended must change, during high-level panel sessions at the third African Media Convention held from 15-17 May in Accra, Ghana.

Academics as drivers of AI

Professor Admire Mare, an associate professor in the department of communication and media, University of Johannesburg in South Africa, told University World News that academics, as researchers, have a major role to play in the development of an AI ecosystem in Africa.

“We need to have our own patents for AI tools and machine-learning systems. We need to train the next generation of AI developers in Africa, and that can be done at the university. Universities must be where we see a lot of AI tools being developed,” he said.

Mare also urged governments across the continent to be intentional in funding the research and development agenda of the university towards AI development, saying that, without this, Africa will continue to depend on AI systems from the Global North, with all their shortcomings – bias, discrimination, among other issues – that come with them.

He said: “One of the key issues now is to drive AI literacy. We need to make sure that, from a young age, our people know what AI means and what coding is all about by inputting our own language systems and data and developing apps that can recognise our languages, cultures and beliefs.

“Government should not just invest in universities, but ensure technology is embedded in what we do and put the right legislation framework in place to encourage investment in the AI ecosystem. We need to ask ourselves: ‘What is African about AI?’ We can’t run away from the fact that natural resources from Africa are powering these technologies, and we are part of this conversation, but unfortunately, we are not part of the conversation as developers and change-makers.

“We are pretty much in the conversation as consumers, and it is a very sad situation. Going forward, we need to develop our own technologies that can preserve our languages, promote our cultures, and help us develop things that can move Africa to the next level,” said Mare.

AI into curriculum

Professor George Nyabuga, an associate dean at the Aga Khan University’s Graduate School of Media and Communications in Nairobi, Kenya, said Africa needs to carry out extensive research and teaching on AI to help people understand what it means, its deployment, and its effects on society.

“Scholarship in terms of research and teaching is critical to people’s understanding of what AI stands for. AI has always been there. It has only become more popular in recent years through apps such as ChatGPT,” he said.

“Another aspect is to distil AI knowledge into curriculum development from primary school to college and then university. That way, we are disseminating, not only information that is critical to people’s understanding, but also knowledge that will ensure those students understand how to apply AI and develop it to respond to African challenges. Technology is never neutral, so we need to think of how to apply it in the African context,” he stated.

Collaboration

To bring AI adaptation in Africa to fruition, collaboration between academia, researchers, AI developers, policymakers – at the level of the African Union and its member states – ministries of science and education, and regional economic communities, is crucial, said Adiatou Fatty, communications lead of the Education, Science, Technology and Innovation, or ESTI, unit at the AU Commission.

“All of these are critical actors in forging a coalition to embed AI in our education system. Africa cannot be left behind in this Fourth Industrial Revolution, marked largely by the transformation of AI. Now, education is defined largely by digitalisation, and when we talk about digitalisation, AI is at the forefront. We cannot afford to be left behind in that stride.

“We also need to be aware of African vulnerabilities when it comes to AI. Devolving AI into learning contents and resources for schools from early childhood to secondary and tertiary level should be our focus, and the AU is trying to come up with an AI strategy. It is also important to provide the necessary infrastructure,” he said.