The most prestigious MBA programs in the world are facing a unique crisis, fuelled by AI innovation. For decades, graduating from Harvard Business School, Stanford, or Wharton was a guarantee of immediate employment and fast career advancement. Yet since 2024, these elite institutions have witnessed their highest unemployment rates in over a decade, with a quarter of Harvard Business School graduates still without employment three months after graduation, followed closely by Stanford at 22% and Wharton at 20%. This represents more than double the unemployment rate from just two years ago, when only 10% of Harvard MBAs remained unemployed after three months.
Enter the AI silicon(e) floor
While economic factors certainly play a role, we are witnessing the emergence of something far more fundamental and concerning: a phenomenon I have named the AI silicon(e) floor. Unlike the theoretical glass ceiling that has historically prevented minorities from ascending to leadership positions, the AI silicon(e) floor suggests a much bigger problem; an invisible barrier that prevents graduates from even setting foot on the first rung of the corporate ladder. This technological barrier is fundamentally different because it does not just limit advancement, it blocks entry entirely and is affecting all people regarding of status, education level or demographic.
Understanding the AI silicon(e) floor
The AI silicon(e) floor represents the technological displacement happening at the bottom of the career ladder, where AI systems are increasingly capable of performing the routine, repetitive tasks typically assigned to new graduates. The data supporting this phenomenon is stark. According to reports, an astounding 32% of entry-level job postings have disappeared (to May 2025) since the launch of ChatGPT in late 2022. This decline outstrips the overall job market reduction, suggesting that entry-level positions are being influenced by AI automation.
Why graduates?
Today’s AI systems currently excel at the assignments traditionally assigned to new graduates: research, data analysis, content creation, and administrative support. Why pay for staff when AI tools can do it better, cheaper, faster? Deep Research tools on OpenAI, Perplexity and Claude (amongst others), compile and analyse information from hundreds of sources to create comprehensive reports at the level of a research analyst… in minutes. Why hire a junior research assistant when an AI can deliver superior results faster and at a fraction of the cost?
Similarly, LLMs can produce high-quality written content, analyse data sets, and even generate strategies in minutes. Therefore, companies globally are asking themselves, why pay more for less?
The children are our future
The answer to this is clear. Today’s graduates are tomorrow’s innovators, leaders, and decision-makers. If we block an entire generation from gaining the foundational workplace experience necessary for a successful career, who will lead the companies of the future? When we all retire, who will take over?
The silicon(e) floor solution
The solution is not to halt AI development, pull the plug or stop innovating. The answer is always to upskill. If the role of the graduate is changing thanks to AI, we must start teaching future graduates the new skills they need. With every industrial revolution comes a skills shift that catapults us into the future. We must quickly understand what this skills shift is and pivot our training to build our workforce of tomorrow.
The end of technical skills as we know it?
Graduates may (debatably) no longer need to become programmers or data scientists, but they absolutely must understand how to leverage AI tools to enhance their productivity and deliver superior results. This means understanding AI capabilities and limitations and developing the critical thinking skills necessary to validate and refine AI-generated outputs.
At this stage, AI systems are not designed to be fully autonomous, and they all work best when humans are in the loop. For example, physicians using AI assistance achieve better diagnostic accuracy than either party working alone. The future (as we currently know it) belongs to humans who can work in harmony with AI, not those who compete against it.
Global responses and educational innovation
Forward-thinking institutions and nations worldwide are beginning to address this challenge. In June 2025, the UK Government recently announced a whopping £187M investment in national AI skills training with a vision of AI literacy for 7.5M workers by 2030. An even more ‘AI-advanced’ nation, Singapore, has a long-standing National AI Strategy which launched in 2017 with gusto and has made them a global leader in AI upskilling.
Educational institutions are also adapting. Arizona State University was the first major educational institution to fully integrate a LLM (ChatGPT) across academics, teaching, research and operations. Many others have since followed suit.
The leadership pipeline crisis
So how do we turn this crisis around? The most effective upskilling programs focus on practical application rather than theoretical understanding. AI upskilling is not an academic issue. It is a skills issue. Our current workforce must quickly be able to put newfound skills into practise before industries globally grind-to-a-halt. No amount of textbook learning will deliver this at the speed or scale needed.
Fundamentally, the implications of the AI silicon(e) floor extend far beyond the career paths of graduates. Organizations depend on a steady pipeline of junior talent who gradually develop the institutional knowledge, leadership skills, and strategic thinking capabilities necessary for senior roles. If AI displaces this entry-level tier, who will run the companies tomorrow?
The end of ‘learning on the job’?
Consider the typical career progression in consulting, finance, or legal sectors and how it’s being affected by AI automation. Junior staff traditionally spend years learning effective research, analysis, and client communication skills while working on increasingly complex projects. This apprenticeship model has been the foundation of professional development for centuries in every country in the world and across all sectors. Now that AI can perform these foundational tasks, how do we ensure the next generation gets the chance to ‘learn on the job’?
Building AI-ready graduates
The solution requires coordinated action across educational institutions, employers, and policymakers. This is not a problem that can be solved alone. Universities must integrate applied AI training into all degree programs, not just technical fields. This includes teaching students how to craft effective prompts, evaluate AI outputs critically, and combining human insight with AI efficiency.
Educational and training providers need to rapidly create industry-specific AI training pathways. Law students need to understand legal AI tools, business students need to master AI-powered analytics, and finance students must learn to leverage AI for efficiency while maintaining data protection.
We do not need to create a world of technical experts, but we must produce graduates who can understand and collaborate with AI systems in their chosen fields.
The employer responsibility
Employers also bear responsibility for re-thinking their recruitment and training strategies. Rather than simply replacing junior roles with AI, forward-thinking organizations must quickly re-design entry-level positions. They must pivot in their understanding and reward employees who blend practical skills and knowledge with integrative AI education. Because these are the valuable employees of tomorrow.
Beyond the technical silo
The most effective AI training and re-training programs share a crucial trait. They are not confined to technical roles or departments. AI is not just a Computer Science. Neither is it just a Data Science. Nor is it a standalone subject reserved for specialists. AI characterises a fundamental shift in how we work, learn, and perform in every job, every sector, and every level of an organization.
Successful organizations and educational institutions are recognizing that AI literacy is as essential as digital literacy was a generation ago. This means moving away from the outdated view that AI is only for coders or data scientists, and toward an understanding that every employee, from HR to marketing to operations, must know how to apply AI tools and concepts in their daily work.
The choice ahead
In 2025 we stand at a critical juncture for the future of our workforce. The AI silicon(e) floor is real and is already affecting the career prospects of thousands of graduates from even the most prestigious institutions. We can either allow the silicon(e) floor to solidify, creating a generation locked out of traditional career pathways, or we can quickly pivot to ensure graduates possess the skills necessary to thrive in an AI-augmented workplace.
The choice is not between humans and AI. It is between education systems that prepare students for yesterday’s job market and those that equip them for tomorrow’s opportunities.
Recent graduates with applied AI skills will find themselves crossing the silicon(e) floor to reach the corporate ladder. Those without such skills may find themselves perpetually slipping on an increasingly slippery surface.
The institutions, employers, and policymakers who act now to bridge this skills gap will create competitive advantages that last for decades. Those who delay will find themselves trying to build leadership pipelines on foundations that no longer exist.
The AI revolution is here. The question is whether we’ll use it to elevate human potential or watch it create barriers that our most talented graduates cannot overcome. And as AI systems grow in intelligence, it will gradually begin to climb the corporate ladder wiping out rung after rung of skilled professionals. Graduates today, middle managers tomorrow. Now is the time to act.
The silicon(e) floor or the corporate ladder, the choice is ours, and the time to choose is now.
About the Author
Tamsin Deasey-Weinstein is a workforce strategist and AI thought-leader who serves as a senior director at the University College of the Cayman Islands (UCCI). A Council member of the Cayman Islands Chamber of Commerce and co-chair of the National AI Education Steering Group, she spearheaded a landmark national study on the future of Cayman’s workforce and leads sector-specific think-tanks translating research into policy. An MIT-certified executive in AI and Machine Learning, Tamsin’s work, featured in Forbes, the BBC and several academic publications, focuses on practical, inclusive upskilling frameworks that align education, industry and government to drive sustainable economic growth.
