VentureBeat presents: AI Unleashed - An exclusive executive event for enterprise data leaders. Network and learn with industry peers. Learn More
New research on inclusivity in tech reveals some concerning results: 84% of all tech employees acknowledge their products are not inclusive, according to a report from the Capgemini Research Institute. What’s more, the firm found stark differences between the opinions of leadership executives and underrepresented employees including women and “ethnic minorities” when it comes to inclusion practices.
When presented with positive statements meant to evaluate equity and inclusivity, executives overwhelmingly felt the statements were true for their organizations, while very few underrepresented employees felt similarly. For example, 85% of executive leaders said they believe their organizations provide equitable opportunities for career development and promotions to every employee, but only 18% of the underrepresented employees agreed. Similar numbers were reported for all seven inclusivity statements posed by the research team. Elizabeth Kiehner, VP of enterprise transformation at Capgemini Invent, told VentureBeat, “That’s an incredibly wide perception gap.”
“These gaps demonstrate that true accountability needs to be established across the majority of organizations,” she said. “There’s an enormous opportunity to focus on inclusivity and appropriately recalibrating in order to create an improved reality for all in this pressing moment.”
The urgency of inclusive tech
Not only are diverse and inclusive teams more profitable, but they’re more likely to create inclusive products. Specifically, organizations with advanced inclusive practices are four times more likely to create inclusive products, according to Capgemini, which evaluated inclusivity by considering an organization’s training, growth opportunities, and to what extent employees feel comfortable offering their perspectives, among other factors.
The tech industry has seen example after example of products and services hitting the market with obvious harms or blind spots. There have been sexist Google features and racist Snapchat filters, for example, but the list could go on and on. Time after time, people wonder, “why didn’t anyone notice this?” Across subfields, technologists are sounding the alarm about the dangers of homogenous teams that can’t and don’t account for the experiences of underrepresented groups. Overall, only 26% of computing-related jobs are held by women; just 3% are held by African American women, 6% by Asian women, and 2% by Hispanic women. And studies show these women, especially women of color, feel invisible at work. To put a number to it, only 16% of women and ethnic minority tech employees surveyed by Capgemini believe they’re well represented in tech teams.
“When tech teams themselves are not inclusive, or when tech team members do not even feel a sense of belonging, the design suffers,” Kiehner said.
And the stakes are only increasing as technology — especially increasingly powerful technologies like AI — become more integrated into our lives. Capgemini found consumers are aware of tech-based discrimination, and most have experienced it. Among women of color, nearly half were offered lower credit facility for certain banking products, according to the report. The use of facial recognition, which is known to misidentify Black people, is already common among law enforcement and is even surging in retail stores. Researchers have also uncovered biases baked into AI-powered corporate interview tools that are costing people jobs. Timnit Gebru, the former co-lead of Google’s ethical AI team, who says she was fired after refusing to rescind research about the risks of deploying large language models, once summed up the issue succinctly in a now often-quoted Facebook post: “I’m not worried about machines taking over the world. I’m worried about groupthink, insularity, and arrogance in the AI community.”
How tech teams can more inclusive
Both Kiehner and the report share some guidance for improving inclusivity: drive fairness in AI systems, keep underrepresented users at the heart of design, reevaluate hiring processes for bias, reduce workplace microaggressions, and so on. In addition to taking such actionable steps, Kiehner says teams need to set “inclusive practices as a standard.” And the report concludes by expressing the urgency of the issue at hand.
But while the above steps are indeed necessary, it’s important to acknowledge that this issue runs deeper than a checklist. The lack of diversity and inclusion in tech — and the product dangers and abuses that can result from it — reflect pervasive racism, sexism, ableism, and other discriminations in our society at large. Building and empowering teams that actually represent our society is the only way forward.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.