Bias out. ROI in: The unexpected win of AI regulation
February 2025. While some companies are frantically reviewing their AI-supported recruiting processes to meet the EU AI Act requirements at the last minute, others are remaining calm. They have used the previous months not only to ensure compliance, but also to fundamentally rethink how technology should be used in recruiting.
Twelve months later, it will become clear that these different approaches have divided the industry into those who truly understand AI in recruiting and those who primarily view it as a technical and/or compliance challenge.
2025: The year of coming of age
The beginning of the year was marked by uncertainty: How do we meet the AI Act requirements? Will we have to shut down our systems? How expensive will that be? But over the course of the year, the most successful companies realized that regulation was not the problem, but rather the solution to the real problem: non-transparent, unfair recruiting processes that overlook potential and reproduce bias.
The pioneers of the HR industry saw the requirements of the EU AI Act as an opportunity. They have not only made their recruiting technology compliant, but also fundamentally improved it. Saint Sass, for example, one of Europe’s most sought-after fashion labels, switched to data-based personality analysis during a critical growth phase and was able to fill five key positions in just four weeks, 50% faster than the startup average, with 50% fewer interviews. Or the CHAPTERS Group, an investment company that specifically builds start-up teams. It uses AI-supported assessments to identify who really brings entrepreneurial thinking and action to the table – and achieved a 25% higher interview-to-hire ratio in 2025.
What these success stories have in common is that they have understood that transparency, traceability, and human oversight are not bureaucratic hurdles, but quality characteristics of excellent recruiting.
The learning curve of the HR industry
2025 was also a year of painful lessons. Companies that had implemented AI tools hastily and without careful consideration had experiences they would rather have avoided: algorithms that reinforced existing biases instead of minimizing them. Black box systems that no one in the company really understood. Candidates who felt devalued by non-transparent processes and withdrew their applications.
The industry has learned that not all AI is the same. The crucial difference lies in the underlying technology and philosophy. While CV parsing systems enable historical patterns and, as a result, often discrimination, language-based personality analysis takes a different approach. It does not evaluate which universities are listed on a resume or which company names are impressive, but rather how people communicate, think, and solve problems. It reveals what traditional methods overlook: intrinsic motivation, teamwork skills, willingness to learn, and entrepreneurial thinking.
The ROI of ethical AI
Responsible AI is not only ethically imperative, it also makes good business sense. By 2025, more and more companies will have recognized this. In times when a bad hire can quickly cost six-figure sums and a shortage of skilled workers is becoming a limiting factor for growth, quality of hire is the decisive metric.
Companies that invested in transparent, explainable AI systems in 2025 report measurable improvements: shorter time-to-hire with higher accuracy, lower turnover in the first twelve months, and more diverse teams without pressure to meet quotas. The ROI comes not from maximum automation, but from better decisions.
The companies that understand this are likely to win the war for talent. Not despite the AI Act requirements, but because of them.
Outlook for 2026: Mastering technology and human understanding
What does this mean for 2026? The fundamental question of “Do we use AI in recruiting?” has been answered. The focus is shifting to “How good is our AI?” Quality standards for recruiting technology will become established, not only through regulation, but also through measurable business impact.
2026 will be the year when AI becomes invisible in recruiting, but in the best sense. It will become as commonplace as email, but its quality will be the decisive differentiating factor. The leading companies will be those that think beyond compliance and use AI strategically: for potential prediction instead of CV screening, for team fit analyses instead of isolated individual assessments, for holistic people analytics instead of stand-alone solutions.
Recruiters will continue to evolve into curating decision-makers who combine data-driven insights with human judgment. Empathy, contextual understanding, and strategic thinking will become more important, not less. The best HR departments will be those that master technology and people skills to an equal degree.
The foundation has been laid
2025 was the year that set the course. Companies that are now on the right side of this development – with transparent, fair, effective AI – have a competitive advantage that will be decisive in 2026. While they are already working on the next generation of recruiting innovation, others are still struggling with the basics.
The year AI came of age in recruiting is also the year recruiting became more human. This apparent paradox is the real innovation: technology at the service of better human decisions.
What a wonderful prospect for the coming year.
Prof. Dr. Florian Feltes
Prof. Dr. Florian Feltes is co-founder and co-CEO of zortify and a forerunner in AI-supported HR innovation. Together with his team, he develops intelligent personality diagnostics and helps companies identify the perfect candidates—without expensive assessments and without bias. His vision: a world in which every company can effortlessly form high-performance teams and create work environments that allow human potential to flourish.
1 in 700 is not enough! – How companies really hire based on skills
Predicting retention: How companies can avoid costly employee turnover
