May 30, 2024

Mitigate HR AI Bias: 5 Crucial Questions from Gartner® to Ask Your Vendors

Mitigate HR AI Bias: 5 Crucial Questions from Gartner® to Ask Your Vendors

Download your copy of the Gartner Mitigate Bias from AI in HR Technology report, compliments of SkyHive by Cornerstone. 

Over half of HR leaders are concerned about potential bias and discrimination from AI, according to Gartner research. Many HR leaders plan to buy AI tech and are monitoring technology providers for compliance and ethics. 

Gartner recommends that HR leaders responsible for technology strategy should:

  • Map possible sources and outputs of bias for each AI use case in HR to assist in flagging areas of risk and monitoring vendors for their commitment to responsible AI practices.
  • Require and evaluate bias mitigation from HR technology providers offering AI functionality by assessing criteria related to their data, algorithms, organizational context, regulation compliance and ethical considerations.
  • Promote transparency into the potential impacts of AI’s bias by collaborating with external and internal stakeholders to take decisive steps in protecting the organization, the future of work and society at large.

Source: Gartner Article, “3 Key Themes Dominate Gartner’s Hype Cycle for HR Technology”, 2 January 2024, <

The consequences of inaction are high. Organizations whose HR tech contracts do not obligate vendors to implement responsible AI are facing increased financial and reputation risk. 

On the plus side, organizations that adopt responsible AI in HR tech achieve greater employee experience and trust in the organization, benefiting employee retention and career development. 

Source: Gartner, Mitigate Bias From AI in HR Technology, Helen Poitevin, 16 October 2023. 

Gartner maps possible sources of AI bias for the following HR use cases:

  1. Attrition rate and flight risk analysis 
  2. Career coach 
  3. HR virtual assistant 
  4. Internal talent marketplace 
  5. Learning personalization 
  6. Performance feedback 
  7. Skills management 
  8. Talent acquisition
  9. Voice of the employee research 
  10. Workforce management 

Gartner recommends asking HR AI vendors the following five questions. We've included an answer for each from the SkyHive perspective. 

1. What data is the vendor solution using? 

SkyHive profiles go beyond resumes, CVs, and LinkedIn profiles to include education, hobbies, and credentials. 

SkyHive brings together comprehensive labor market data in real time. Our data sources include online professional profiles, academic journals, government data, curriculum documentation, online training content, patent applications, and more. 

SkyHive Labor Market Intelligence (LMI) provides global data on job vacancies and in-demand skills in real time. SkyHive ingests over 28TB of raw data a day across more than 150 countries and territories. 

Our data collection, processing, and privacy protections follow ethical AI practices. The following six principles define our approach to ethical and responsible AI: transparency, explainability, robustness, trust, confidentiality, and accountability. 

The SkyHive approach is skill extraction. We produce the market’s largest knowledge graph of human capital data. This includes over 5 billion job descriptions and 1 billion anonymized job profiles. The LMI looks at patent applications and academic journals to derive new skills that are becoming in demand. For more details on SkyHive data sources, visit SkyHive’s FAQ page

SkyHive has the largest and most flexible skills ontology. Our ontology provides the ability to extract, interpret, and normalize skills, as well as their relative proficiency. 

When people self-report on their skills, they impose limitations. They don’t fully realize how skilled they are. On average, individuals identify 11 skills for their particular role. Using SkyHive, that number jumps to an average of 34. Identifying a wider range of skills, and adjacent skills available to learn, has important benefits for professional development, performance management, learning and development.

In awarding SkyHive one of the “Next Big Technologies Working for Social Good in 2023”, Fast Company explains that SkyHive “uses AI to match people to jobs they might not have thought were a fit but that they actually have the transferable skills for…. and it even shows learning opportunities that could help bridge a skill gap to enter a new career.”

Gainwell Talent and Development Manager Julie Moore describes how once Gainwell employees began seeing ideas for their career paths in SkyHive, “People started really getting the word out. [Asking their peers] ‘Oh, you haven’t done that yet? You can see different paths in the labor market.’ That really helped people add more skills.” 

Gainwell’s initial launch with SkyHive reached 10,600 enrolled employees with an average of 22 skills per profile. At that time, employees had already completed more than 600 courses.  

Several months later, Gainwell employees had 36 skills on average per profile. Employees completed 1,567 courses and 257 employees found mentors. (Skyhive video “Boosting Employee Engagement with Skills-based Career Paths” with Julie Moore at Gainwell Technologies).

For more customer examples visit

2. What assumptions go into the vendor’s algorithms to create a “match”?

SkyHive allows organizations to have visibility into the degree of skills match between an individual and an open role. This skills-focused approach allows the matching without the potential biases inherent in traditional methods.

Many organizations are revising degree requirements when hiring, and emphasizing experience and skills instead. Skills are increasingly developed outside of traditional university education including self-taught online classes, employer education, and side gigs.

SkyHive employs different bias mitigation methods depending on the situation, including for machine translation, sentiment analysis, language models, and word embedding analogies.

We employ:

  • Algorithms for discrimination discovery.
  • Discrimination prevention by means of fairness-aware data.
  • Human in the loop approach.

The framework for fairness-aware modeling includes the following:

  • Baseline: Train a model on all available input variables in the source dataset, including protected attributes. 
  • Remove Protected Attribute: Train a model on input variables without protected attributes. This is what’s referred to as a naive fairness-aware approach. 
  • Relabel Target Variable: Train a model using the Relabeling fairness-aware method. 
  • Counterfactually Fair Model: Train a model using the Additive Counterfactually Fair method. 
  • Reject-option Classification: Train a model using the Reject-option Classification method.

3. How tailored is the vendor’s AI use case to the organizational context?

More than 100 large enterprises and governments have selected SkyHive to empower the transition from jobs to skills. SkyHive tailors our offerings for each organization and industry (enterprise, government, educational, and non-profit organizations) to enable strategic workforce planning, improve internal mobility, and help reskill and upskill their workforce. 

We train our AI using only publicly available data, and never using customer data. 

4. What approaches does the vendor use to mitigate the harms caused by bias? 

Biased data may contain for example gender, years of experience, and job titles. SkyHive takes the approach of creating a dataset and allowing the model to measure bias. SkyHive then subtracts the effect of that bias on the outcome.  

From a technical perspective, SkyHive uses the XGBoost method with Shapley Additive Explanation (SHAP) to extract feature contributions at individual prediction levels. 

Extreme Gradient Boosting (XGBoost) is a machine-learning algorithm that uses decision trees to solve supervised learning problems. 

SHAP values are a way to explain the output of a machine learning model. SHAP calculates a value that represents the contribution of each feature to the model outcome. These values are then used to understand the importance of each feature and to explain the result of the model in ways people can understand. 

5. How will the vendor comply with current and future regulations? 

SkyHive is at the forefront to take a proactive approach for ethical and responsible AI. SkyHive Skill Models are Armilla Verified as free of AI bias. SkyHive meets the standards set by the EU AI Act, New York City’s Local Law 144, and future regulations. We never use your data for training our AI.  

To learn more, download your complimentary copy of the Gartner Mitigate Bias from AI in HR Technology report.


Gartner Article, “3 Key Themes Dominate Gartner’s Hype Cycle for HR Technology”, 2 January 2024, <>.

Gartner, Mitigate Bias From AI in HR Technology, Helen Poitevin, 16 October 2023. 

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.

About SkyHive by Cornerstone

Recognized by the World Economic Forum for our contributions to the future of work, SkyHive enables organizations and communities worldwide from Best Buy, the Government of Canada, and Collège La Cité to Unilever and Zinnia transition from jobs to skills, backed by partnerships including Accenture, SAP, and Workday and non-profits Opportunity@Work and JobsFirstNYC. Learn more about how to unleash human potential at

Cornerstone OnDemand Inc. acquired SkyHive on May 22, 2024. See the press release for the exciting news. Cornerstone Galaxy, the complete AI-powered Workforce Agility platform, allows organizations to identify skills gaps and development opportunities, retain and engage top talent, and 
provide multi-modal learning experiences to meet the diverse needs of the modern workforce. Learn more at

Download PDF

Ready to unleash potential across your workforce?

The world's most ethical Al people technology to help you transition from jobs-based to skills-based — award-winning, demonstrated, and internationally recognized.

Request a demo

Related resources

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.