Liji Thomas’ Post

Let's Keep Beauty Real. Even in Al? A lot of my previous work & research centered around gender bias in #AI and AI fairness. It’s evident that AI not only reflects but amplifies stereotypes and unrealistic beauty standards - affecting not just women. Exposure to AI output influences all of us heavily. Dove's recent campaign highlights these issues, and I've been asked frequently for my thoughts: 💡 Yes, shedding a light on this problem is important. We need more big corporate players to do so! It’s a starting point. 🎥 About the Campaign: Although the video feels empowering, it is not based on a real fine-tuned fair and inclusive model or retrieval-augmented generation (RAG). Moving forward, I'd love to see such models in action, shifting from superficial brand campaigns to meaningful, long-term changes in how AI portrays women. 💁🏽♀️ Dove must have huge image set of diverse models - why not train a diverse AI model on that? Alternatively, partner with The Algorithmic Justice League or other institutions to transform this initiative into a sustainable, reusable tool in #AI. In my opinion, this video feels good to watch but real change requires going beyond the surface! Video ©️Dove

Artificial Intelligence (AI) is clearly a hot topic in the world today, and while it is an exciting technology, it is important to pay attention to the ethical and economic implications of AI moving forward. A recent report showed that over one-third of hours worked in the US could be automated by 2030, and there is worry that this will disproportionally affect women. Take a look at this article to learn more about this topic and solutions to make sure men and women are affected the same.

"How Businesses Can Mitigate Bias in AI and Find Balance," - Saleh Seyedzadeh at The Data Lab - Innovation Centre delves into the critical issue of AI bias and its implications for diversity and inclusion within organizations. Drawing from Harnham's comprehensive Diversity in Data Guide for research, the piece sheds light on the consequences of neglecting diversity in data sources and AI development teams (📩 https://lnkd.in/gBwFkWnk). It emphasizes the stark reality of bias in AI systems, as evidenced by studies like the Gender Shades project and clinical algorithm discrepancies, highlighting the urgent need for a diverse workforce to combat these biases effectively. The exploration into the challenges and solutions around AI bias underscores the importance of varied data sets and the inclusion of diverse perspectives in creating more equitable AI models ⤵ #AIEthics #DiversityInTech #BiasInAI #InclusiveAI #Harnham #Rockborne

"Usher in an era where the adoption of AI acts as a lifting and equalizing force" -> An inspirational goal in this recent Harvard Business Review article. While enjoying all the great posts on how AI can contribute and lift L&D, I am missing a similar discussion and sharing of best practice on how the use of AI in L&D can also be an equalising force. I am afraid that this may mean that we miss an important opportunity to be a force for good as L&D/ HRD. Questions that come to mind, inspired by Patrizia Zanoni and the invite for the International Workshop "Unpacking how new technologies are reshaping diversity and (in)equality at work" in May 2023: - How does the use of AI affect how L&D is being reorganised and how does this reorganisation relate to processes of inclusion/exclusion and (in)equality? - How does the use of AI in L&D redefine diverse workers’ learning opportunities? To which extent does the use of AI in L&D entrench past inequalities (think f.e. the gender digital divide)? - Which role does AI in L&D play in the production of prescriptions of ideal employees and ideal performance? With which effects on historically subordinated groups of workers? - Under which conditions can the use of AI in L&D reduce discrimination and biases? - How can AI support L&D teams in understanding and engaging with diversity, equality and inclusion in their work? Love to hear your thoughts and any literature/ posts that you can recommend! #genderequityintraining #learninganddevelopment #diversityequityandinclusion

Great report about representation of Women in AI from Deloitte. This is great info. Strongly encourage you to read! https://lnkd.in/g6NZsKzS. From the infographic: https://lnkd.in/gVeJx6CZ. Q: Why is it important for women to be represented in AI? The tools that use AI and ML often express their power by emulating human decision-making, perception, or discernment? A. How well can they work if the perspectives of only some humans, and not others, go into their creation? #womeninAI #AI #artificialintelligence #womenindatascience #womenintech #representationmatters

“Only by making other voices heard, can we ensure a more fair and equitable development of data science and AI, and the products and services that are built on it.” – Sara Colombo #Feminist AI can transform the way we approach the #AI industry, by challenging the notion that a few, typically affluent, Western, male individuals should have control over the way AI is developed, designed, and deployed. The TU Delft | AI Initiative interviewed Sara Colombo, the co-founder and co-director of the Feminist Generative AI Lab, which is a joint research lab between Delft University of Technology and Erasmus University Rotterdam. The lab serves as a pioneering hub where generative AI, design, and data feminism intersect. As a discipline and as a practice, #datascience has historically been led by a narrow cohort of people, representing a homogeneous demographic. This lack of diversity has contributed to the (often unconscious) perpetration of biases and stereotypes, which pose significant risks to society. 📢 Increasing #diversity in data science and AI not only mitigates these risks, but also fosters creativity and produces better outcomes. Read the story here: https://lnkd.in/dn45QekZ

Repost from @bloombergbusiness • While artificial intelligence is seeding upheaval across the workforce, from screenwriters to financial advisors, the technology will disproportionately replace jobs typically held by women, according to human resources analytics firm Revelio Labs. Revelio Labs identified jobs that are most likely to be replaced by AI based on a study by the National Bureau of Economic Research. They then identified the gender breakdown of those jobs and found that many of them are generally held by women, such as bill and account collectors, payroll clerks and executive secretaries. Advances in AI are aggravating gender disparity in the global workforce, where companies are considering cutting some staff and leveraging generative AI in their workflow. @bloombergequality 📷: Jason Alden/Bloomberg

As a discipline and as a practice, #datascience has historically been led by a narrow cohort of people, representing a homogeneous demographic. This lack of diversity has contributed to the (often unconscious) perpetration of #biases and stereotypes, which pose significant risks to society. Increasing #diversity in data science and AI not only mitigates these risks, but also fosters creativity and produces better outcomes. 🌍🤝    ♀️ #Feminist AI can transform the way we approach the #AI industry, by challenging the notion that a few, typically affluent, Western, male individuals should have control over the way AI is developed, designed, and deployed. In honor of #InternationalWomensDay2024, the AI Initiative interviewed Sara Colombo, who leads the Feminist Generative AI Lab at TU Delft and advocates for other voices in data science and AI. “Only by making other voices heard, can we ensure a more fair and equitable development of data science and AI, and the products and services that are built on it.” 🗣️ Read the story here: https://lnkd.in/dn45QekZ

On March 27th, I'll be joining A.Team and a panel of amazing women to discuss the bias in AI. There's 100 in-person spots and 500 virtual so RSVP soon if you're interested! https://lnkd.in/etSmYi8K #ai #artificialintelligence #machinelearning #technology #business #nyc #responsibletech #kleinerpowell #juliettepowell