Ethical AI: Everything HR Professionals Should Know About the Proposed EU AI Act
August 8, 2023
Filippa Oldenburg
Imagine pressing the fast-forward button in a movie, zooming ahead to a future where hiring has evolved into an exciting blend of technology and ethics. Enter the spotlight: the EU AI Act, a major player transforming how we hire. Let's explore its current proposal and its impact on the future of hiring.
The act made its debut in 2022 and is set to be fully implemented by 2025. But this isn't just another regulation. It's a game-changer, guiding HR professionals into a world where AI and responsibility are solidified.
Discovering the EU AI Act: A Fresh Start for Hiring

The European Union's suggested AI rules mark a significant step toward managing the use of artificial intelligence in different areas, including the complex world of recruitment. These rules showcase the EU's dedication to ethical AI and aim to find a careful balance between making the most of AI's great possibilities while also protecting essential rights.

Breaking Down the Act’s Core Elements

The EU AI Act introduces a tiered approach to AI applications, categorizing them into three distinct levels of risk: unacceptable risk, high risk, and low or minimal risk. When it comes to recruitment, our focus is primarily on high-risk AI applications, which includes systems that assess applicants' qualifications and suitability for specific job positions. Let's break down the high-risk AI in recruitment and explore its implications.

Defining High-Risk AI in Recruitment

Imagine a scenario where an AI system is tasked with automatically shortlisting candidates for a critical job opening. This system relies on advanced algorithms to analyze vast amounts of data, from educational backgrounds to work experience, to determine the best-fit candidates. The EU AI Act identifies these types of systems as high-risk AI in recruitment, as they have a significant impact on individuals' chances of employment. To ensure ethical and fair utilization of such systems, the regulation mandates several important requirements.

  • Transparency: Consider a situation where you are a job applicant interacting with an AI-driven recruitment process. The EU AI Act ensures that you are informed about the AI's involvement in your assessment and that you have the right to access information about how the AI system operates.
  • Human Oversight: Think of AI as a co-pilot rather than the sole decision-maker. The regulation emphasizes the importance of human experts being involved in the design, development, and ongoing monitoring of high-risk AI systems. Their presence aims to help mitigate bias and discrimination, ensuring that the system's outcomes are just and equitable.
  • Data Quality and Bias Mitigation: Imagine a diverse pool of candidates with varying backgrounds and experiences. The regulation calls for measures to ensure the quality and integrity of the training data used by high-risk AI systems. This helps minimize the potential for biased outcomes that could disproportionately impact certain groups.
Impact on the Recruitment Process

The EU AI Act paves the way for a recruitment process that is transparent, accountable, and fair. Employers, guided by ethical AI principles, must offer candidates a clear understanding of AI's role in selection. Companies using high-risk AI systems are responsible for any negative results, promoting responsible AI adoption that values fairness and diversity.

Preparing for EU AI Act: HR Professionals' To-Do List

As HR professionals, preparing for the upcoming EU AI Act requires a proactive approach. Here's a practical task list to help you navigate the changes ahead:

  • Learn about the EU AI Act: Study the EU AI Act thoroughly. Understand its main rules, like how AI applications are categorized and the guidelines for high-risk AI systems in recruitment.
  • Assess Your AI Usage: Check if your hiring methods involve high-risk AI systems. Examine how these systems automatically review and select candidates.
  • Ensure Transparency: Make sure your AI-driven hiring processes are transparent. Candidates should know AI is being used and should have access to details about how the technology functions.
  • Involve Experts: Include human experts in creating, developing, and overseeing your AI systems. This helps prevent biases and ensures fair evaluations.
  • Check Data Quality: Examine your training data for AI systems. Confirm it's of good quality and unbiased. Take steps to reduce the chance of unfair outcomes.
  • Enhance Candidate Experience: Improve how candidates interact with your hiring process. Clearly explain AI's role and its impact on decision-making.
  • Review Legal Compliance: Work with your legal team to ensure your AI practices match the requirements of the EU AI Act.
  • Allocate Resources: Set aside resources to meet the new obligations, including making your processes transparent and ensuring human oversight.

Our Commitment to Ethical AI: Hubert's Role

In the evolving AI landscape, ethical solutions like Hubert are on the rise. Hubert, our AI tool for unbiased interview screening, minimizes bias and ensures diverse assessments. It promotes inclusivity by providing equal opportunities to all candidates. Hubert prioritizes transparency — candidates are aware that their initial interview involves AI and can receive feedback afterward. Crucially, the tool doesn't make final hiring decisions; instead, it empowers human decision-makers. This collaboration strikes a balance between AI efficiency and human judgment. The tool showcases AI's potential for an improved, fairer, and efficient recruitment process, highlighting the potential harmonious coexistence of humans and technology.

Recognizing Ethical Excellence: Hubert's "Ethical AI" Award

In June 2023, our dedication to ethical AI practices was acknowledged when we received the "Ethical AI" award from ManpowerGroup during VivaTech in Paris. This recognition underscores our commitment to advancing responsible AI in recruitment.

Conclusion: Embracing an Ethical Path Forward in Recruitment FAQs

The recruitment landscape undergoes a profound shift due to the upcoming EU AI Act. This regulation emphasizes transparency, accountability, and fairness in AI use, influencing HR practices towards an unbiased candidate selection future and a better experience for all. However, what the final version will look like remains to be seen.

Implementation period

  1. Proposal and Debate (2022-2023): The Act is introduced, and experts discuss and refine it.
  2. Evaluation and Refinement (2023-2024): Member states assess the Act, making improvements.
  3. Approval and Adoption (2024-2025): The Act is approved by the European Parliament and Council.
  4. Transition Period (2025-2026): Organizations adjust to align with the Act's provisions.
  5. Full Implementation (2026 and Beyond): The Act's rules are fully integrated into AI practices.

FAQs

What is the EU AI Act?

The EU AI Act is a comprehensive set of rules proposed by the European Union to regulate the use of artificial intelligence across various sectors, including recruitment.


How does the EU AI Act impact recruitment?

The regulation introduces a tiered approach to AI applications, with specific requirements for high-risk AI systems used in recruitment. It emphasizes transparency, human oversight, and bias mitigation.


What is Hubert?

Hubert seamlessly integrates into your ATS as an interview screening tool, streamlining and enhancing high-volume recruitment processes through ethical AI interviews. Hubert not only accelerates the process but also improves decision-making accuracy and increases the overall quality of hires, all while reducing unconscious bias and upholding ethical standards.


How does Hubert reduce bias in recruitment?

Our structured interviews and algorithm ensure that each candidate is treated and evaluated in the same, anonymous manner. This contributes to reducing bias in the candidate selection process.


How do the EU AI Act and Hubert work together?

The EU AI Act establishes rules for responsible AI use, and Hubert exemplifies ethical AI principles in practice. This combination of human interaction and AI efficiency ensures fair and transparent recruitment practices, where bias is drastically reduced.

Just like any developing laws, it's important to mention that the EU AI Act discussed here is currently a proposal and hasn't become an official law yet. Although our approach might be in line with the ethical suggestions of the proposed EU AI Act, the final version of the law could still undergo modifications.

Insight
Ethical AI: Everything HR Professionals Should Know About the Proposed EU AI Act
August 8, 2023
Filippa Oldenburg
Contact
Give us a call
General inquiries
hello@hubert.ai
Swedish office
Malmskillnadsgatan 32
111 51 Stockholm, Sweden