New Study Release!

[Virtual Assistant]

New Study Release!

We’re excited to share that the pilot study "Enduring Disparities in the Workplace: A Study in the AI Community" is now live on arXiv! This cross-group collaboration explores the persistent gaps and experiences of marginalized groups working in AI.Key Findings Include:

• Enduring disparities in workplace experiences for underrepresented and/or marginalized subgroups
  
• Accessibility remains an important challenge for a positive work environment, and disabled employees have a worse workplace experience than their non-disabled colleagues
  
• The implementation of DEI initiatives may differ from their perceived impact on the workplace
  
• Patterns of microaggressions in AI/ML workplaces reflect those of broader society
  
• Intersectional groups face compounded disadvantages that are not captured by examining single identity dimensions alone
  

Access the full study here: https://arxiv.org/abs/2506.04305We extend a heartfelt thank you to all members of the community who took the survey and to all the researchers who helped us design and distribute it, your participation made this research possible. We also appreciate the transparency shown by employees at DeepMind, Google Research, and Apple, which enabled the reporting of disaggregated workplace experiences from companies with 30+ respondents.This was truly a collective effort led by multiple affinity groups, including Women in Machine Learning (@WiMLworkshop), Black in AI (@black_in_ai), Queer in AI (@queerinai), LatinX in AI (@LatinXinAI), and Deep Learning Indaba (@DeepIndaba).We’d love for you to read it, share your thoughts, and spread the word. Feel free to tag us on Twitter/X (@WiMLworkshop), BlueSky (wimlworkshop.bsky.social‬), or LinkedIn, we’d love to reshare your content and keep the conversation going.

Thank you again to everyone who contributed to this important step forward in understanding and addressing inequity in the AI workplace