jobs (see Exhibits 6, 7, and 8).35 Women were also more likely than men to leave the industry, which further narrowed the pool of women eligible for promotions or leadership positions (see Exhibit 9)?"El Research suggested, however, that the low number of women and underrepresented minorities who studied and pursued computer science was likely driven as much by unconscious bias as it was by conscious choice. Research showed how stereotypes associating men with math and science may have prevented young women from pursuing STEM degrees.\" Similarly, teachers tended to be less likely to encourage female, Black, and Latino students from pursuing STEM subjects due to internalized stereotypes that white male students were better suited for STEM.33 Notably, computing had not always been male-centric; most of the first software programmers in the 19505 and 19605 were women}9 Only as programming grew increasingly protable did men become overrepresented.'10 A third school of thought held that cultural issues in tech had created a hostile workplace environment for women and underrepresented minorities, which made them more likely to leave the industry.41 Underrepresented minorities in tech were more likely to report \"unfairness" as the motivating factor for leaving a job compared to their white or Asian peers, and almost 90% of female tech workers reported experiencing some form of gender-based discrimination in the workplace.'12 Still others attributed the lack of diversity in the industry to the belief that success in tech depended not on collaboration and teamwork, but rather \"almost entirely on innate genius.\"43 Schmidt called brilliant employees who exhibited a contrarian streak \"aberrant geniuses," explaining to V'fired: \"You need these aberrant geniuses because they're the ones that drive, in most cases, the product excellence.\"4 Research had shown that due to stereotypes, people tended to associate the term \"genius\" more frequently with men, making it harder for women to gain respect in the tech field.\" This belief in the lone tech genius sometimes made it easier for companies to justify tolerating bad behavior by so-called geniuses because they were viewed as innately gifted and thus irreplaceable.'16 To increase diversity, Google had formed special recruiting teams and partnered with external organizations such as Black Girls Code, Lesbians Who Tech, and the National Center for Women in Technology." The company had also partnered with Howard University, a renowned historically Black university, to recruit computer science graduates, and implemented programs like BOLD, an internship for underrepresented minorities.48 Google's desire to diversify went beyond ethical concerns about fairness. An unrepresentative tech workforce was less likely to develop products that resonated with the broadest swathe of the population.49 And at least one study found that diverse companies were more nancially successful than their less-diverse peers.50 Google's lack of diversity also had legal repercussions. A years-long Department of Labor (DoL) investigation into the company ultimately reported \" systemic compensation disparities against women pretty much across the entire workforce.\" 51 DoL representatives further claimed that \" discrimination against women in Google is quite extreme, even in [tech]."52 In January 2017, the DoL sued Google after the company refused to release its historic compensation data.53 Several months later, three female former employees also sued Google for gender discrimination, claiming that they were hired at lower compensation levels than their male colleagues with similar qualifications.54 Google denied the allegations of gender pay discrimination, arguing that the company used an objective evaluation system for determining employee compensation.55 Google subsequently conducted an internal investigation on gender pay equity, and claimed that the company was actually underpaying many men, not women; Google raised the salaries of thousands of male employees as a result. 56 Critics, however, pointed out that Google's findings did not evaluate whether the company hired women at lower salary bands than their male peers.\