As artificial intelligence (AI) gains momentum, concerns about fairness and bias have come to the forefront of discussions. Komal Singh, a senior product manager at Google, acknowledges the complexities of addressing fairness in AI but believes that society has a second chance to rectify biases within this rapidly evolving technology.
AI is hailed as a transformative technology with the potential to disrupt various industries, offering efficiency, expedience, and solutions to complex global challenges.
While some embrace AI's potential, figures like Elon Musk and Steve Wozniak caution that advancements are occurring too rapidly, necessitating safeguards against unintended consequences.
Geoffrey Hinton, a prominent computer scientist often dubbed the 'godfather of AI,' left Google to freely discuss AI's dangers. He has cited concerns including bias, discrimination, job displacement, echo chambers, fake news, battle robots, and existential risks associated with AI.
Singh highlights instances where AI models return biased results, such as associating certain roles or images with specific genders or races.
To mitigate racial biases, Google collaborated with Harvard University's Dr. Ellis Monk to create an open-source skin tone scale that enhances AI model accuracy in recognizing darker skin tones. This scale is already integrated into Google's camera detection technology, diversifying image search results.
Google's "Media Understanding for Social Exploration" (MUSE) project studied 12 years of American television shows, revealing trends in screen time given to different skin tones.
Tech giants are fiercely competing to develop advanced AI technology, with investments in the billions. In Canada, generative AI is projected to boost the economy significantly.
AI is being used in real-world scenarios to enhance breast cancer detection, sequence genomes, streamline e-commerce search experiences, and create digital supply chain platforms.
Google is committed to adhering to principles that guide its AI development, including a promise not to create AI technologies that harm or violate human rights.
As AI continues to advance, society has a unique opportunity to rectify biases and challenges within the technology. Google, alongside other tech giants, is actively working to address fairness issues and promote responsible AI development. Balancing AI's potential for transformation with ethical considerations remains a critical endeavor as the technology evolves.
