TL;DR:
- Google’s CEO emphasizes responsible AI development as a top priority for 2024.
- Google’s internal AI ethics oversight team, RESIN, undergoes leadership change and restructuring.
- RESIN played a pivotal role in evaluating projects for adherence to Google’s AI principles.
- Founder Jen Gennai steps down, raising questions about the team’s future.
- Team members were divided into two groups, with some moving to the trust and safety division.
- Google remains tight-lipped about the future of AI principles reviews.
- The restructuring aims to strengthen Google’s commitment to responsible AI development.
- Concerns arise over the potential impact on product development and consumer trust.
- Google has been streamlining operations to accelerate AI advancements.
- The industry’s focus on assessing societal risks related to advanced AI remains paramount.
Main AI News:
In a recent communication to its workforce, Google CEO Sundar Pichai underscored the company’s commitment to responsible AI development as its top priority for 2024. However, concerns have emerged among employees about Google’s ability to uphold this commitment, following significant changes within its primary internal AI ethics oversight team. According to insiders, this small but pivotal team, known as Google’s Responsible Innovation team or RESIN, has undergone a leadership change and restructuring.
RESIN was originally housed within the Office of Compliance and Integrity, part of the company’s global affairs division. Its primary role was to evaluate internal projects for alignment with Google’s AI principles, which define the guidelines for the development and use of AI technology. This role has become increasingly crucial as Google competes in the rapidly evolving field of generative AI. In the past year alone, RESIN conducted over 500 reviews, including one for the Bard chatbot, as disclosed in Google’s recent annual report on AI principles work.
The uncertainty surrounding RESIN’s future became apparent when its founder and leader, Jen Gennai, stepped down from her role as director of responsible innovation. Gennai’s LinkedIn profile now lists her as an AI ethics and compliance adviser at Google, leading many to speculate about her imminent departure. Google divided Gennai’s team of approximately 30 members into two groups, with some remaining within RESIN and others transferring to the trust and safety division, responsible for combatting abuse of Google services. Fortunately, no layoffs have been reported, although it remains unclear how the responsibilities and AI principles reviews will be distributed going forward.
When approached for details, Google spokesperson Brian Gabriel declined to elaborate on how RESIN’s AI project reviews will be managed in the future. However, he characterized the restructuring as a sign of Google’s unwavering commitment to responsible AI development. Gabriel stated, “This restructuring has placed our Responsible AI team at the heart of our well-established trust and safety efforts, which are integral to our product reviews and strategies. It will enable us to reinforce and expand our responsible innovation initiatives throughout the organization.”
While Google has a history of organizational changes, RESIN has remained relatively untouched since its inception. Despite numerous other teams and hundreds of personnel dedicated to AI oversight within Google, RESIN occupied a unique position as the most prominent and all-encompassing team responsible for upholding AI principles across the company.
In addition to Jen Gennai’s departure, RESIN also lost one of its influential members, Sara Tangdall, a specialist in AI principles ethics. Tangdall is now the responsible AI product director at Salesforce, as indicated by her LinkedIn profile. She chose not to comment on the matter, and attempts to reach Gennai for comment went unanswered.
Google’s Responsible Innovation team was established in 2018, following public protests within the company against Project Maven, a Pentagon contract that utilized Google algorithms for analyzing drone surveillance data. RESIN played a vital role in implementing a set of AI principles aimed at ensuring that Google’s AI technology benefits humanity and is never used for weapons or in violation of human rights. Notably, Jen Gennai played a significant role in formulating these principles.
RESIN allowed teams across Google to submit their projects for evaluation, offering feedback and occasionally blocking ideas that contradicted AI principles. This included preventing the release of AI image generators and voice synthesis algorithms that could be exploited to create deepfakes.
While seeking guidance from AI principles was not mandatory for most teams, Gennai emphasized that early AI system reviews paid off by preventing costly ethical breaches. She stated during a Google conference in 2022, “Responsible AI, when implemented properly, enhances product quality by identifying and mitigating the harm caused by unfair bias, enhancing transparency, and bolstering security.”
In 2022, RESIN was incorporated into the global affairs division’s compliance unit, with Google describing this move as centralizing governance across all product areas. However, some team members feared that this change could shift RESIN’s focus towards protecting Google’s interests rather than safeguarding consumers.
As Google faces increasing competition in the AI landscape and an evolving consumer landscape with alternative platforms like TikTok and ChatGPT, employees have raised concerns about the potential consequences of hastened product development, exacerbated by the restructuring of RESIN.
Google has been streamlining its operations over the past year to accelerate the delivery of advancements to users and focus on key AI initiatives. This includes discontinuing services like the Podcasts app and scaling back features in Google Assistant. The company has also faced challenges in its ad sales, influenced by new regulations and court orders on privacy and anticompetitive behavior in the post-pandemic economy.
Pichai and other Google leaders have affirmed their commitment to responsible AI development while accelerating its progress. Google’s participation in a voluntary White House pledge, alongside OpenAI, Microsoft, and other major AI developers, reflects the industry’s recognition of the need to assess societal risks and national security concerns related to advanced AI technologies.
Conclusion:
Google’s restructuring of its AI ethics oversight, particularly the RESIN team, reflects the company’s commitment to responsible AI development. However, it raises questions about the future direction of AI principles reviews and their impact on product development. As Google continues to streamline its operations and navigate challenges in the market, the industry should closely monitor how these changes shape its AI initiatives and ethical standards.
