In the global pursuit of crafting regulatory frameworks for artificial intelligence (AI), tech giant Google has put forth a proposal for a risk-based approach. Rather than applying uniform rules across all AI applications, Google suggests that regulations should be tailored to the level of risk associated with different uses of AI. Pandu Nayak, the Vice President of Search at Google, emphasized the need for a proportional approach, asserting that regulations should be focused on the application level rather than the overarching technology.

Speaking at the Global Partnership on Artificial Intelligence (GPAI) Summit, Nayak highlighted the importance of assessing actual risks associated with specific AI applications. He argued against a "one-size-fits-all" strategy, noting the varying risks in different sectors. Nayak emphasized that in sectors such as agriculture, the risks from AI usage differ significantly from those in other domains.

One focal point of concern raised during the discussion was the issue of bias in AI systems. Responding to queries about addressing bias in search results and Google's generative AI platform Bard, Nayak underscored ongoing research efforts and the importance of training models on reliable datasets. Google aims to mitigate bias by avoiding training on unsafe content. The conversation also delved into the government's response to bias in AI-generated content, with Nayak expressing support for a risk-based approach aligned with existing laws governing biased content.

Additionally, Nayak highlighted the role of cross-border trusted data flows in mitigating bias in generative AI platforms. By enabling diverse demographic representation in training data through cross-border data flows, Google believes it can contribute to addressing bias concerns. This perspective aligns with the government's strategy, as Minister of State for Electronics and IT Rajeev Chandrasekhar recently emphasized the importance of sharing public data only with proven and trusted sources in the AI industry. Nayak expressed support for this approach, aligning Google's stance with the government's initiative to foster access to datasets for trusted AI platforms.