Citibank just made headlines by requiring all 175,000 of its employees globally to complete AI training within 60 days. Every single employee, from entry-level staff to senior executives, must finish the program called "Asking Smart Questions, Prompting Like a Pro." This isn't a suggestion or recommendation. It's required.
This mandate isn't an isolated decision by one forward-thinking bank. JP Morgan now requires AI training for all new hires as part of standard onboarding. Bank of America reports over 90% of its workforce uses AI tools after completing training. Wells Fargo has trained thousands through Stanford's human-centered AI curriculum. PwC has integrated AI literacy into company culture through gamified monthly tournaments and quizzes.
The trend extends well beyond finance. The EU AI Act, effective February 2025, now legally mandates AI competency training for all employees who interact with or operate AI systems. Over 250 CEOs from companies like Adobe, AMD, American Airlines, and Cognizant signed an open letter urging mandatory AI education in all K-12 US schools.
The shift is undeniable. AI literacy is transitioning from "nice to have" to baseline competency. Organizations are treating AI fluency the same way they treat cybersecurity awareness or workplace safety—as something every employee needs, not an optional skill for technical teams.
Citibank's program uses adaptive learning technology. Employees with existing AI experience can complete it in about 10 minutes. Those new to AI get approximately 30 minutes of instruction. The program covers fundamental concepts: what AI is, how to interact with it effectively, basic prompting techniques. Nothing extraordinarily complex, but comprehensive enough to ensure baseline literacy across a massive global workforce.
JP Morgan takes a different approach, embedding AI training directly into new hire onboarding. Before employees start their actual roles, they complete modules on AI fundamentals and how the bank uses these tools. The message to new employees is immediate: AI literacy is expected here from day one.
Bank of America's success metric tells an important story. Over 90% of their workforce now uses AI tools in their daily work. That adoption rate didn't happen by accident or through gentle encouragement. It followed comprehensive training that gave employees both permission and capability to integrate AI into their workflows.
Wells Fargo partnered with Stanford's human-centered AI program, bringing academic rigor to their corporate training. PwC gamified the experience with monthly tournaments and quizzes, turning AI literacy into a cultural norm rather than a compliance checkbox. Different approaches, same underlying principle: everyone participates.
These programs share common elements. They're mandatory. They're time-bound. They establish clear minimum standards for AI competency across entire organizations.
The EU AI Act represents a significant shift in how governments are approaching AI literacy. Effective February 2025, it legally requires companies to ensure personnel acquire sufficient AI knowledge, including technical understanding and context of use. This applies to all employees who interact with or operate AI systems—which in modern workplaces means nearly everyone.
Even US-based organizations should pay attention. Regulatory trends often follow similar patterns across developed economies. What starts as European compliance frequently becomes global best practice, then American standard. The GDPR followed this trajectory. Climate disclosure requirements are following it now. AI literacy requirements appear headed the same direction.
Beyond government mandates, the private sector is mobilizing independently. Over 250 CEOs signed an open letter pushing for mandatory AI education in all K-12 schools. They're not waiting for today's workers to catch up—they're ensuring tomorrow's workforce arrives AI-literate. The signatories include major employers across industries, from technology to transportation to professional services.
Leading AI training firms report surging demand for customized mandatory learning paths, role-based corporate workshops, and compliance certification programs. Organizations are building AI literacy infrastructure the way they built digital literacy programs a generation ago. The question isn't whether to build this infrastructure. It's how quickly you can deploy it.
Real-world patterns from organizations using AI training platforms show clear trends. Data from Sidecar's AI Learning Hub, which serves associations specifically, reveals a stark difference in outcomes based on how leadership frames training requirements.
When leadership makes training mandatory, participation rates consistently reach near-universal levels. Organizations see their teams not just completing courses but earning certifications and applying what they learn. When training remains optional or voluntary, participation typically falls into the 20-30% range. In some cases, it drops even lower despite the training being available to everyone at no additional cost.
The correlation between leadership stance and participation outcomes holds across different organization sizes and types. The pattern repeats itself consistently: mandatory requirements drive broad participation, optional approaches reach only a fraction of the team.
Optional training means the majority of your team doesn't participate. They remain unprepared for a workplace transformation that's already underway. Those participation gaps don't close over time—they typically widen as early adopters pull further ahead and non-participants fall further behind.
The data suggests something important about human behavior and organizational change. People respond to signals about what matters. When leadership says training is important and encouraged, some people engage. When leadership says training is required and everyone does it, nearly everyone engages. The difference between encouragement and requirement determines whether you transform your entire organization or just a small slice of early adopters.
Associations often operate differently than corporations when it comes to staff directives. Many association leaders have built careers in environments that value consensus-building. Volunteer governance structures can influence how staff decisions get made. Top-down mandates might feel at odds with the collaborative culture many associations have carefully cultivated over years or decades.
There may be concerns about how mandates affect staff morale or organizational culture. Some leaders worry about appearing heavy-handed with requirements. The preference for voluntary participation often comes from genuinely valuing staff autonomy. These are reasonable considerations rooted in how associations have successfully operated.
The committee-based decision making that works well for many association initiatives can slow down responses to rapid technological change. The same collaborative approach that builds buy-in for strategic plans can create bottlenecks when speed matters. And right now, speed matters.
Associations also tend to have flatter organizational structures than corporations. The traditional corporate hierarchy makes mandates easier to implement—leadership decides, management executes, employees comply. Association culture often resists that model, preferring influence over authority, persuasion over directive.
These cultural differences are real and worthy of respect. But they create tension when facing technological disruption that doesn't care about organizational culture or preferred decision-making styles. AI is transforming work whether association leaders feel comfortable mandating training or not.
Here's a straightforward question every leader should answer: Do you believe AI will significantly affect the skills required in your workplace? Most leaders, if being honest, would answer yes. The technology is already reshaping how work gets done across industries and roles.
If the answer is yes, then preparing your workforce isn't optional. It's fundamental leadership responsibility. Your job as a leader, perhaps your most important job, is to grow people. That's the real product you're making in your organization. Whether you run a software company, professional services firm, or association, you're in the business of developing people.
Serving your employees' and volunteers' futures requires preparing them for AI. This isn't just about your organization's operational needs. The person who leaves your organization in five years needs AI literacy to thrive in their next role. That's your responsibility today.
The data proves the point. Optional training leaves most people behind. A leader who chooses comfort over effectiveness—who prefers to encourage rather than require—is making a choice about whose future matters. The 20-30% who self-select into optional training will be fine. They'll develop AI skills, apply them to their work, and build valuable capabilities. The 70-80% who don't participate? They're being left unprepared for a transformed workplace.
That's where the language of malpractice fits. Medical malpractice happens when a doctor fails to provide the standard of care that a reasonable professional would provide. Leadership malpractice happens when a leader fails to provide the development that a reasonable leader would provide. In 2025, with AI reshaping every profession, reasonable leaders are ensuring their teams develop AI literacy.
You can disagree about methods, timing, or specific approaches. But the fundamental premise is hard to dispute: leaders who don't prepare their teams for AI are not serving those teams well.
The word "mandatory" can feel heavy-handed. It conjures images of rigid requirements and punitive consequences. But look at what mandatory actually means in successful AI training programs.
Citibank gives employees 60 days to complete training that takes 10-30 minutes. That's not unreasonable. That's barely more restrictive than asking people to update their passwords quarterly or complete annual compliance training. The requirement comes with ample time and reasonable expectations.
You can't literally force anyone to do anything. People always have choices. The practical meaning of "mandatory" in this context is: "optional to the extent that you don't want to work here anymore." If people want to remain employed, they complete the AI training. If they choose not to complete it, they're choosing not to meet baseline job requirements.
This is identical to how organizations treat other baseline competencies. Nobody questions mandatory cybersecurity training. Workplace safety training is required, not suggested. Compliance training is non-negotiable. Organizations routinely mandate certain knowledge and behaviors because they're essential for operations, legal compliance, or risk management.
AI literacy deserves the same treatment as these other foundational skills. The technology is reshaping work as fundamentally as computers and the internet did in previous generations. Treating AI training as optional makes as much sense as treating email training as optional in 1998.
Mandatory doesn't mean elaborate. Citi's program takes under an hour for most employees. You can start with free courses from Google, Microsoft, OpenAI, or Anthropic. You can use specialized programs like Sidecar's AI Learning Hub that provide association-specific context and role-based learning paths. Multiple good options exist at various price points.
Most organizations should mandate two things at minimum:
Beyond that baseline, offer electives:
Make these available and encouraged, but they don't need to be universally mandatory.
Some roles may require additional mandatory training beyond the baseline. If someone's job involves deploying AI tools for members, they need deeper expertise than someone in a role that only uses AI occasionally. Calibrate requirements to actual job needs.
AI is also blurring traditional role boundaries in interesting ways. Marketing people can now do things that used to require developers. Finance staff can perform analyses that previously needed data scientists. Cross-pollination across departments becomes more valuable as AI expands what's possible for everyone. Encouraging people to take courses outside their immediate role can unlock unexpected capabilities.
One-time training establishes baseline competency. Continuous learning builds real expertise. The most effective approach combines mandatory initial training with ongoing opportunities to deepen knowledge.
Block 15 minutes daily on your calendar for AI learning. Make it a recurring appointment with yourself. Morning, evening, lunch break—whenever works, just make it consistent. Use that time to try new AI tools, experiment with different prompting techniques, read about recent developments, or take additional courses.
The compounding effect of this habit is remarkable. Fifteen minutes daily doesn't sound like much. Over a week, that's 75 minutes. Over a month, five hours. Over a year, 65 hours of focused learning. That volume of consistent practice moves you from novice to knowledgeable to expert to world-class within a couple years.
The habit matters more than the intensity. Marathon study sessions have their place, but daily practice creates lasting behavioral change. Organizations should encourage this habit alongside the mandatory baseline training. Create channels for sharing what people learn. Recognize employees who consistently expand their AI capabilities. Build a culture where continuous learning is expected and celebrated.
Citibank is training 175,000 employees. JP Morgan made it part of onboarding. The EU made it law. Two hundred and fifty CEOs are pushing for K-12 AI education. The evidence is overwhelming: mandatory training works, optional training fails.
Association leaders can choose comfort—optional training, 20-30% participation, most of the team left behind. Or they can choose effectiveness—mandatory training, near-100% participation, everyone prepared for the transformed workplace ahead.
The malpractice isn't about being too aggressive with mandates or moving too fast or disrupting comfortable organizational culture. The malpractice is leaving 70-80% of your team unprepared for their professional futures. That's the actual choice.
This is a leadership moment that requires courage over comfort. The organizations succeeding with AI are making training non-negotiable. They're not asking for volunteers. They're not hoping people self-select into preparation. They're requiring it, tracking it, and ensuring everyone participates.
Your turn.