The potential of Generative AI in content creation is vast, and Rwandans are actively embracing these transformative tools. Artificial Intelligence encompasses a diverse array of technologies aimed at enabling machines to mimic human cognitive abilities, from perception to interpretation and learning. Within this spectrum lies Generative AI (Gen AI), which includes sophisticated large language models capable of generating diverse content, spanning from text to images, through extensive data training. ALSO READ: Mental health: Can AI-powered chatbots enhance access to care, combat stigma? A January 2024 report by the International Monetary Fund (IMF), titled ‘Gen-AI: Artificial Intelligence and the Future of Work’highlights the rapid evolution of AI, particularly with the emergence of Gen AI, expanding its potential applications. This evolution suggests an impact on reshaping job roles and labour dynamics. Furthermore, the IMF report underscores the importance of prioritising digital infrastructure and human capital investment in low-income countries and emerging markets. “With such investments, AI could help alleviate skill shortages, expand the provision of healthcare and education, and improve productivity and competitiveness in new sectors.” How is Generative AI being utilised? The utilisation of Gen AI in 2024 marks a leap forward in content creation processes. With tools like ChatGPT, creators can effortlessly craft scripts, while Voice Mod enables voice modulation, transforming recordings to emulate the voices of public figures like Donald Trump or Morgan Freeman. Subsequent video editing is streamlined with Adobe Project Fast Fill, facilitating seamless publication across social media platforms. AI analytics further revolutionise content dissemination, employing machine learning algorithms to automate data analysis tasks previously undertaken by data analysts. ALSO READ: Rwanda to host Africa’s inaugural high-level summit on AI It’s evident that AI is revolutionising content creation, creators can now devote less time to mundane tasks and channel their energy into finding inspiration and weaving compelling narratives. “Automated processes have long been integral to our creative endeavours, with tools like Adobe Photoshop, Premiere Pro, Illustration, and After Effects shaping our workflow. However, the emergence of Artificial Intelligence is simplifying these processes further, reducing the necessity for extensive expertise in traditional editing software,” said Shebbah Tanganyika, an experienced content creator managing YouTube channels since 2021 who attested to the impact of AI tools on workflow efficiency. “Previously, a week’s worth of editing now takes no more than a day, thanks to AI-powered tools. Beyond efficiency, these tools also foster creativity, offering a departure from conventional editing approaches. Given the demand for frequent uploads, what I would describe as ‘fast food’ content, tools like 'Filmora' with AI-powered features have become indispensable. “Among the standout features we use mostly is AI Copilot Editing, which analyses videos and provides tailored recommendations during editing. Additionally, AI Text-To-Video fills gaps in footage by generating non-copyrighted videos based on written prompts, offering endless possibilities. Audio refinement can also be streamlined using tools like Adobe Podcast AI, particularly beneficial for voice recordings captured on phones, often plagued by background noise,” she said. ALSO READ: Impact of Generative AI on music, content creation Tanganyika emphasised that creatives not embracing AI tools are missing out on streamlining their operations. “To enhance its appeal further, mastering these tools doesn’t demand extensive learning because countless tutorial videos are readily available online, offering accessible guidance for anyone eager to learn.” Challenges and potential Audace Niyonkuru, CEO of Digital Umuganda, a Rwanda-based AI and open data company dedicated to improving access to information and services in local African languages, played a role in developing the largest open voice dataset in African languages. “The challenge lies in establishing regulations to ensure responsible use of AI models and fair compensation for the artistic works used in their training. While tools like OpenAI’s Sora and Google’s Lumiere hold promise for various content creation tasks, including image-to-video conversion and stylised generation, they also introduce new risks. Regulatory frameworks must evolve alongside this technology to prevent misuse. For instance, using someone’s content to train AI models without consent should entail penalties. The rise of deepfakes also poses concerns, perpetuating misinformation through the creation of realistic yet fabricated content.” On Wednesday, February 21, hundreds of professionals signed an open letter advocating for stringent regulation of AI-generated impersonations, commonly known as deepfakes. Although the immediate impact on legislation remains uncertain, this collective action serves as an indicator of expert sentiment on this controversial issue. The letter, endorsed by over 800 individuals within and adjacent to the AI sector at the time of publication, underscores the pressing concern posed by deepfakes and calls for government intervention to address their proliferation effectively. Presently, deepfakes frequently involve the manipulation of imagery for purposes such as sexual exploitation, fraudulent activities, or dissemination of political misinformation. While initially confined to doctored images, the emergence of generative AI has fuelled the proliferation of deepfake videos, primarily circulated on pornographic platforms. A 2019 study by the cybersecurity firm Deeptrace revealed that 96 per cent of online deepfake video content constituted non-consensual pornography, the targets of such malicious manipulation have included public figures like Taylor Swift, Margot Robbie, Barack Obama, Donald Trump, Jenna Ortega, among others. “These risks necessitate intervention through proper regulations and laws to prevent misuse of these tools by creators, thereby safeguarding against issues like copyright infringement. It is imperative for creators to transparently indicate their use of AI-generated content. Developing countries like Rwanda must implement financial penalties to incentivise responsible AI usage,” Niyonkuru said. “In Rwanda’s context, there is a pressing need for tools to combat misinformation, given our historical context. Content moderation and fact-checking tools are essential to address issues such as hate speech, especially since current detection systems are predominantly owned by large corporations. Increased awareness of AI threats among creators and consumers alike is paramount, and while the majority of AI models originate from Western countries, Africa must also take proactive steps.” He added: “Infrastructure development is crucial, along with the creation of tailored data sets for our continent. At Digital Umuganda, we prioritise data collection, recognising the potential implications of Western reliance on African data. Societal biases inherent in AI models are why Africa should be leveraging this technology to avoid becoming mere consumers in the fourth industrial revolution. “Rwanda’s investment in AI technology is notable among African countries, and we have started attracting foreign direct investments. However, building public trust is essential for widespread AI adoption. Strengthening regulatory capacity to align with global standards and best practices will foster public confidence and facilitate the creation and consumption of AI-driven content.” Is Rwanda set for AI adoption? To gauge Rwanda’s readiness for embracing AI, Victor Muvunyi, Senior Technologist in Emerging Technologies at the Ministry of ICT and Innovation, shared insights. Muvunyi’s lecture formed an integral part of a training session tailored for Rwandan journalists keen on reporting in this domain. “Preparedness levels vary significantly, encompassing foundational AI readiness, which includes robust digital infrastructure and a skilled workforce capable of AI adoption, as well as second-generation preparedness involving innovation and legal frameworks. “In economies with substantial exposure to AI and well-established foundational readiness, particularly advanced and certain emerging market economies, there’s a need to prioritise enhancing digital innovation capabilities and refining legal and ethical frameworks to govern AI advancements. This includes bolstering regulatory frameworks crucial for fostering societal trust in AI tools, alongside fostering innovation and seamless integration,” he said. “Addressing cybersecurity risks is integral within regulatory frameworks. Initiatives such as the establishment of the Personal Data Protection and Privacy Law aim to mitigate data breaches and privacy concerns, essential for nurturing trust, safeguarding personal information, and upholding data privacy principles, pivotal for delivering social and economic services. “Effective preparedness for AI adoption is paramount to leverage its potential while mitigating associated risks. Historical precedents underscore the role of a country’s preparedness level in maximising AI benefits while managing downsides. The National Artificial Intelligence Policy serves as a foundational roadmap empowering Rwanda to harness AI’s benefits while navigating its risks. Aligned with Vision 2050, Smart Rwanda Master Plan, and other key national strategies, this policy positions Rwanda as a leading African Innovation Hub and Center of Excellence in Artificial Intelligence by rallying local, regional, and international stakeholders,” Muvunyi added. The advent of Generative AI is poised to revolutionise the global economy, drawing parallels with a new industrial revolution. Its consequences for economies and societies remain hard to foresee and nowhere is this uncertainty more visible than in labour markets, where AI holds the promise of bolstering productivity while concurrently raising concerns about job displacement and augmentation. As industry experts tout AI’s potential, the question persists; are they indeed prophetic? Moreover, how are you embracing this burgeoning technology?
