Alliance Magazine: Philanthropy needs to invest in building trust in AI
By Ujjwal Gupta and Millie ARORA
Within a mere six months of its launch, ChatGPT skyrocketed to become the 17th most visited website globally, surpassing established platforms like Reddit, LinkedIn, and Netflix. This rapid adoption is just the tip of the iceberg – the AI market is projected to reach an astonishing $400 billion in revenues by 2027, with forecasts suggesting it could boost the US GDP by 21 percent and create 97 million new jobs by 2030.
Global philanthropic leaders are embracing AI, arguably faster than any previous technology. The Gates Foundation expanded its Global Grand Challenges grant pool by 66 percent, spurred by the entrepreneurial interest in AI-driven solutions and their potential impact on global health outcomes while the Patrick J. McGovern Foundation and the Novo Nordisk Foundation have each invested over $60 million in AI infrastructure and solutions.
Current philanthropic investments in AI are primarily centred around two fundamental trends: open-source models and fine-tuning existing large language models. Open-source AI models are being heralded as the way forward, rapidly closing the performance gap with proprietary models while offering unmatched levels of transparency. Simultaneously, fine-tuning these models to build specialized AI solutions on the foundational success of existing large language models, such as GPT-4, is considered best practice to avoid reinventing the wheel and optimize efficiency.
However, a crucial consideration is at risk of being overlooked: the importance of building people’s trust in AI.“For philanthropy looking to invest in AI solutions for social impact, leaders should consider developing a community-centric perspective on what it will take to build people’s trust in AI.”
Read the full article on Alliance Magazine.