Microsoft is leveraging AI to enhance inclusivity in its workplace.
By published
Microsoft's head of diversity asserts that enhancing workforce diversity can mitigate biases in artificial intelligence.
At the start of 2023, Microsoft faced significant public relations challenges as it sought to showcase advancements in AI following a substantial investment in OpenAI, the creator of ChatGPT. The integration of an AI chatbot into its Bing search engine positioned Microsoft among the early tech giants to incorporate AI into core products. However, the rollout quickly encountered issues, particularly after a journalist reported a disconcerting interaction with Bing. Users began sharing alarming screenshots that suggested the chatbot was using offensive language and making bizarre claims about world domination. In response, Microsoft swiftly implemented restrictions on the AI's functionalities. Subsequently, the company transitioned to a new chatbot, Copilot, which is now part of Microsoft 365 and Windows.
Microsoft is not alone in grappling with AI-related controversies. Critics argue that the incident highlights a broader negligence regarding the risks associated with AI in the tech sector. For instance, a notable error during a live demonstration resulted in a significant loss in market value. The AI model, now known as Gemini, later faced criticism for perceived bias after it appeared reluctant to generate images of white individuals for specific requests.
Despite these challenges, Microsoft maintains that AI can foster equity and representation if developed with appropriate safeguards. One proposed solution is to enhance the diversity and inclusivity of the teams responsible for creating the technology. Lindsay-Rae McIntyre, Microsoft's chief diversity officer, emphasizes the importance of building inclusive AI and technology for the future. With over two decades of experience in human resources within the tech industry, McIntyre advocates for embedding inclusive practices in AI research and development to ensure better representation throughout the organization.
This focus on diversity is crucial, especially as AI products have revitalized Microsoft's nearly 50-year-old brand. In July, the company reported a 15% increase in annual revenue, reaching $64.7 billion, largely driven by growth in its Azure cloud services, which have thrived amid the AI surge.
To create AI that is empathetic, relevant, and accurate, McIntyre argues for the necessity of training models with input from a diverse array of developers, engineers, and researchers. However, this approach does not guarantee a solution. The large language models that power tools like Copilot and Gemini are trained on vast datasets sourced from the internet. If biases exist within this training data, they are likely to manifest in the AI's outputs, raising concerns as these technologies are deployed in real-world scenarios.
While companies are increasingly adopting AI for hiring processes, experts caution that these tools can perpetuate existing racial and gender biases in candidate selection. Research revealed that AI systems categorized certain professions as feminine or masculine based on historical stereotypes, further complicating the issue of bias in AI.
Nonetheless, Microsoft believes that AI can promote diversity and inclusion if these principles are integrated into the models from the outset. The company asserts that inclusivity has always been a core aspect of its culture, as evidenced by initiatives like the Xbox Adaptive Controller and various accessibility features in Microsoft 365. However, the rapid evolution of AI necessitates a renewed commitment to diversity within the organization.
According to its 2023 diversity report, Microsoft reported that approximately 54.8% of its core workforce comprises racial and ethnic minorities, aligning with industry peers. However, the percentage of women in Microsoft's workforce stands at 31.2%, a slight decline compared to its competitors.
In an interview, McIntyre discussed Microsoft's strategies for addressing biases in generative AI, emphasizing the company's commitment to education on bias and inclusion across all operations. Microsoft is investing in research to identify and mitigate fairness-related harms and is innovating ways to proactively test its AI systems, as outlined in its Responsible AI Standard. The company collaborates with experts from various fields, including anthropology and linguistics, to enhance the perspectives of its engineers and developers.
Despite facing backlash over reports of cuts to its diversity and inclusion team, Microsoft clarified that its commitment to these initiatives remains strong and is expanding. McIntyre highlighted the importance of inclusive technology and the responsibility Microsoft has to its stakeholders to lead in diversity and inclusion efforts.
Another area of focus for Microsoft is making AI more accessible in multiple languages. Operating in a non-native language can hinder productivity and authenticity, presenting an opportunity for AI to better accommodate global linguistic diversity. Additionally, Microsoft has partnered with organizations supporting LGBTQ+ communities to implement features like adding pronouns to user profiles in Microsoft 365, fostering a sense of recognition and care.
To ensure that its workforce of 230,000 understands allyship in culturally relevant ways, Microsoft tailors its strategies to local contexts. For instance, in India, the company conducted a diversity and inclusion experience addressing race, ethnicity, and religion for its leaders. Similar initiatives are underway in Australia and New Zealand for indigenous communities, as well as in the Middle East and Africa, where support for employees experiencing menopause is being prioritized.
Microsoft recognizes the importance of cultural context in AI experiences, aiming to create empathetic technology that resonates with users. The company operates in 190 countries and has established nine global employee-led resource groups to promote an inclusive workplace. With over 5 billion chats and images generated through Copilot and more than 60,000 Azure AI cloud customers, Microsoft is committed to ensuring that its talent keeps pace with the rapid evolution of AI.
To prevent talent from being left behind in the AI landscape, Microsoft has developed an AI learning hub that provides access to the latest insights and training. Employee resource groups are also facilitating upskilling initiatives, enabling employees across various roles to enhance their understanding of AI technology. The company is deploying AI skilling courses to ensure all employees share a foundational knowledge of AI as it becomes increasingly integrated into their work.
For organizations looking to incorporate AI into their HR practices, McIntyre suggests three strategies: forming a community of learning with experts, engaging with AI tools like Copilot to explore their potential for fostering inclusion, and adopting a human-centered design approach that prioritizes empathy and user experience.
More news
