The AI transformation within recruitment is accelerating rapidly. What initially started as experimentation with chatbots and automation tools is now evolving into a structural shift in how candidates search, match and apply for jobs.
According to Geert-Jan Waasdorp of Intelligence Group, the industry is underestimating how quickly candidates are adopting AI driven job search behaviour. While many organisations are still focused on selecting tools, candidates are already building personalised AI agents that automatically search, filter and recommend jobs based on individual preferences.
The candidate is already ahead
Recent labour market data shows that AI adoption among job seekers is growing rapidly. In 2024, around 6% of the Dutch workforce used AI during job orientation. In 2025, that figure has already exceeded 11%, with significantly higher adoption rates in sectors such as IT.
Within technology related professions, more than one in five job switchers already uses AI as a primary source for discovering new career opportunities. The expectation is that AI driven job discovery will soon become one of the top recruitment channels globally.
This shift has major implications for traditional job boards and recruitment platforms. Candidates increasingly expect highly personalised matching experiences that go beyond job titles and keyword searches.
AI matching is becoming more contextual
One of the most important developments is the rise of contextual matching. Modern Large Language Models no longer match solely on skills or job titles, but also on preferences, commuting distance, salary expectations, working hours, culture fit and lifestyle considerations.
According to Waasdorp, this fundamentally changes candidate expectations. AI driven matching increasingly delivers more relevant results than many traditional Applicant Tracking Systems.
As a result, organisations need to rethink how they position vacancies, employer branding and labour market communication inside AI ecosystems such as ChatGPT, Claude and Gemini.
The rise of Generative Engine Optimization
This new reality is also creating the emergence of Generative Engine Optimization, often referred to as GEO.
Just as SEO previously determined visibility within Google, GEO focuses on how organisations become visible inside Large Language Models. When candidates ask AI assistants to recommend the best employers or vacancies in a certain region, the underlying models decide which organisations surface first.
That means recruitment marketing increasingly depends on structured, accessible and trustworthy content that AI systems can interpret correctly.
AI models are evolving faster than organisations
The technological acceleration behind these developments is significant. New AI models continue to improve rapidly in reasoning, writing, coding and task execution capabilities.
At the same time, organisations often struggle to keep up. Research from MIT suggests that a large percentage of AI pilots fail because companies focus too heavily on isolated process improvements rather than broader transformation.
Many organisations still approach AI as a collection of separate tools, while the real opportunity lies in integrated ecosystems built around people, workflows and connected data.
Why data ownership matters
According to Waasdorp, proprietary data is becoming one of the most valuable assets in recruitment.
Organisations that depend entirely on external platforms or suppliers risk losing strategic control over their recruitment intelligence. Companies therefore increasingly need to build their own talent data environments and combine internal insights with external labour market information.
The real value emerges when organisations can transform data into actionable insights, recommendations and automated execution.
MCP may become the universal layer
Another major development is the growing role of MCP, or Model Context Protocol. This technology functions as a universal connection layer between systems, platforms and AI models.
Instead of building expensive direct integrations between every individual system, MCP enables AI models to access multiple data sources simultaneously through a shared protocol.
This could significantly reduce fragmentation between HR, recruitment, procurement and workforce management systems.
What recruitment may look like by 2030
Waasdorp predicts that the recruitment ecosystem could look fundamentally different within the next few years.
Personal AI agents may become standard for both employers and candidates. Traditional ATS systems could gradually lose relevance, while automated skills based matching becomes the dominant recruitment model.
Identity verification systems such as the EU ID Wallet may also reduce fake applications and improve trust within digital hiring processes.
At the same time, recruiters are unlikely to disappear entirely. Human judgement, relationship building and strategic decision making will remain essential, even as AI handles more operational tasks.
Six lessons for recruitment leaders
The AI roadmap for recruitment in 2026 ultimately comes down to several core lessons.
Start with people rather than processes. Focus on candidate behaviour and user experience before implementing technology.
Invest in proprietary data and connected ecosystems instead of isolated tools.
Think beyond short term point solutions, since many standalone AI tools may quickly become outdated.
Shift performance measurement from cost per hire towards productivity and long term talent impact.
Understand the strategic importance of MCP and interoperability between systems.
Most importantly, continuously experiment with AI. The technology is evolving rapidly, and organisations that delay adoption risk falling behind both competitors and candidates themselves.
The central message is clear: the recruitment industry is not waiting for perfect regulation or complete certainty. The transformation is already happening, and candidates are leading the way.
