On Tuesday, Google announced a revision to its Artificial Intelligence (AI) Principles, outlining a refreshed perspective on the company’s approach to AI technology. Previously, the Mountain View-based tech giant identified four key areas where it would refrain from designing or deploying AI solutions, including weapons, surveillance technologies, and tools that could cause significant harm or infringe upon human rights. The updated document has now eliminated this entire section, suggesting a possible shift in the company’s stance towards these previously prohibited domains.
Revamped AI Principles from Google
Google initially unveiled its AI Principles in 2018, at a time when artificial intelligence was still emerging in various sectors. Over the years, the company has periodically revised the document, with the constraints surrounding dangerous applications remaining largely intact until this recent update. The recent changes, however, have sparked discussions about the company’s intentions following the removal of the restrictions.
Evidence of the prior commitments can be found on an archived version of the web page from last week, which included a section titled “Applications we will not pursue.” This section articulated specific prohibitions, including technologies that would “cause or are likely to cause overall harm,” as well as weaponry and corresponding technologies that would directly result in harm to individuals.
Furthermore, the prior guidelines included a pledge to refrain from employing AI in surveillance technologies that breach global norms, along with any systems that would circumvent international law and endanger human rights. The absence of these directives from the revised document has raised concerns regarding Google’s potential venture into these sensitive areas.
In a related blog post, Demis Hassabis, Co-Founder and CEO of Google DeepMind, along with James Manyika, Senior Vice President for Technology and Society, elaborated on the motivations behind the revision.
The executives noted several factors contributing to the update, including the rapid evolution of the AI field, escalating competition, and the intricate geopolitical environment. They emphasized the belief that democracies should spearhead AI development, anchored by principles of freedom, equality, and respect for human rights. The executives further suggested that stakeholders who align with these values must collaborate to ensure AI serves the interests of individuals, fosters global advancement, and supports national security.
