As AI sweeps away old boundaries, bold leadership, ethics and accountability are critical, says IBM’s Deema Alathel
Irresistible waves of change are sweeping through business and society as the AI revolution gathers pace. Legacy systems, long-established organizational habits and entire business models are being washed away as new models begin to take shape. For leaders, the challenge is no longer whether AI will transform their organizations, but how – and how they can navigate the change.
Few people are better positioned to make sense of this moment than Deema Alathel, executive director of global strategic initiatives at IBM. A technologist by training and a strategist by profession, Alathel sits at the intersection of engineering, policy and executive decision-making. Her work spans governments and corporations, national systems and global markets. And she is direct about what she sees going wrong.
Most executives, she says, treat AI as a project. That is the first mistake. “Boards want quick ROIs, proof of concept, something to present,” Alathel explains. “But AI isn’t a project that delivers returns in a quarter or two. The whole organization needs to change how it operates.”
This is not consultant shorthand. Alathel holds a PhD in computer science focused on AI from George Washington University. She’s written the algorithms and built the models – and watched them break. That technical grounding underpins her understanding of the pain points that are common across industries today, as leaders globally grapple with the challenges of adopting AI – and it is the context for her analysis of why so many organizations are struggling to make the progress they expect.
Her central argument is both simple and challenging. AI adoption is not just a technology project – yet most leaders do not understand that they are playing an entirely new game.
The discipline of uncertainty
Alathel’s thinking about leadership is inseparable from her experience as a technologist. “AI forces you to confront complexity, uncertainty and unintended consequences head-on,” she says. “Models break. Assumptions fail. Data reflects human bias. That training taught me humility as a leader. I’m skeptical of hype, and I’m allergic to shortcuts.”
Executives would do well to note that skepticism – especially when their professional expertise lies in non-technological fields. “Most leaders and board members are not technical,” she observes. “They treat this as another IT initiative when the entire operating model needs to shift.”
The result is a mismatch between expectations and reality. Boards push for fast returns, while underestimating the scale of organizational change required – yet they are also hesitant to commit to larger scale investment.
That hesitation is understandable, says Alathel, but costly. “Leaders are reluctant to invest early because they don’t know if it’s worth it,” she points out. “But if you don’t invest now, by the time you act, everyone else will have moved ahead. Progress is accelerating.”
Yet even large organizations are earlier in their journeys than they’d like to admit. Benchmarking against competitors is nigh-on impossible when secrecy surrounds AI investments, but the idea that everyone else has already figured it out is a myth, she says. “Most organizations are only scratching the surface and running pilots. We hear about big-bang projects, but that approach never works.”
Instead, she advocates starting small. Low-risk applications – such as HR processes or document summarization – can help organizations build confidence and capability. “You get people comfortable with the change,” she says. “Then you accelerate from there.”
From local to global
Alathel’s perspective has been shaped by a career that has rapidly expanded in scope. She began in academia, working at her alma mater, King Saud University in Riyadh, before joining IBM Saudi Arabia in 2019. From there, she moved into regional roles spanning the Middle East and Europe, and later relocated to IBM’s New York headquarters with a global remit.
“I started grounded locally in Saudi Arabia, working with national institutions and understanding what it takes to build capability within a specific context,” she reflects. “That progression taught me that effective technology leadership cannot be designed in isolation. It must translate across borders and hold up under global scrutiny.”
Her leadership philosophy did not emerge from any single mentor, she says, generous though many mentors have been during the different stages of career. Instead, she describes learning through pattern recognition across contexts – watching how power is exercised, how decisions are made, and how organizations behave under pressure. “The leaders who shaped me created space for disagreement, protected teams from political noise, and understood that power is borrowed temporarily, not owned,” she says.
Credibility and consequence
The question of who wields that power is a crucial one for many employers – even industries – globally that continue to have an under-representation of women in senior leadership roles.
Working in male-dominated environments taught Alathel specific lessons. “It sharpened me rather than softened me,” she says. “I learned that credibility is built through consistency, not visibility. I learned to speak precisely, decide deliberately, and avoid performative leadership.”
She’s more concerned with systemic questions than representation metrics – particularly those that relate to bias in AI systems. “You need to understand the target audience for a model and reflect them in the team building it,” she explains. Sex, age, regional background and heritage all matter. It’s a matter of managing risk – and it needs to be dealt with up front, not treated as a tick-box exercise in a project’s later stages.
“These things influence outcomes. They get into models without you knowing it when people write the algorithms. That’s why it needs leaders’ attention early.”
The focus on systems rather than symbolism extends to the broader debate about AI’s societal impact, which is poorly framed, she argues. “The conversation is often cast in extremes: salvation or catastrophe. Both narratives are lazy,” says Alathel. “The real opportunity lies in something more uncomfortable: redesigning how organizations think, decide and take responsibility.”
Decision velocity and accountability
For Alathel, AI’s value proposition goes far beyond cost reduction. “Its real value emerges when it reshapes decision velocity and decision quality,” she argues. Organizations that use AI well shorten feedback loops and replace intuition with evidence. “That requires leaders willing to be contradicted
by machines.”
At a societal level, however, the stakes are higher. Context matters. “In environments with strong institutions, AI can improve access and efficiency,” she says. “In weak environments, it could deepen inequality and increase opacity. That’s why leadership matters more than algorithms.”
This is the critical insight. As organizations redesign systems and processes, leaders must stay focused on responsibility, being consistently explicit about who owns decisions and outcomes. “You cannot blame bad outcomes on AI models. You need governance, explainability and human oversight. Those aren’t compliance exercises, they’re leadership choices,” says Alathel. Leaders have to ensure key questions are answered. “Who decides on the team composition? Who ensures model explainability? Who’s responsible for cleansing training data?”
The pace of adaptation is critical. “The biggest risk isn’t that AI moves too fast; it’s that leadership thinking moves too slowly,” she warns. “Many organizations are trying to apply 20th-century management models to 21st-century systems.”
Today’s leaders, she argues, must be comfortable with probabilistic thinking, cross-functional fluency and moral clarity. “I cannot stress that enough,” Alathel says. “Leaders must be able to say not just what can be built, but what should not be deployed, and why.”
Translation and execution
Alathel describes her current role in deliberately practical terms. “I translate between governments and corporations, between engineers and policymakers, between ambition and execution,” she says.
That translation involves building partnerships across borders while guiding executives through complex AI journeys. “Successful leaders invest time in governance models, shared accountability and talent development long before technology decisions are finalized,” she explains. “This is unglamorous work, but it’s where real impact is created.”
In her view, the organizations that stand out are not those with the biggest AI labs. “They’re those with the courage to rethink operating models. They treat AI as organizational capability, not a digital product. And they invest in people.”
The human dimension, we reflect, is sometimes overlooked. The challenge is multilayered. For employers, building internal capability is essential; no organization can simply rely on consultants to build an AI system and then hand over the keys without developing in-house expertise.
National leadership also matters. Developing talent for the future economy requires coordination across institutions. “We need ecosystems that bring together universities, ministries, governments and the private sector,” she argues. The galvanizing focus, she explains, should be a rigorous analysis of skills needs over the next three to five years.
Saudi Arabia’s AI and digital strategy – part of the Kingdom’s national strategy, Vision 2030 –offers lessons for emerging economies, points out Alathel. “It stands out because it’s being built with intent rather than retrofitted,” she says. “There’s rare alignment between policy, capital and talent development. This creates an opportunity not just to adopt AI, but to shape its norms.”
Indeed, developing nations may have unexpected advantages when it comes to harnessing the extraordinary power of AI and other new technologies. “When they’re building infrastructure for these technologies, they’re starting fresh. They’re not slowed down by fixing old mistakes.” From building physical infrastructure like data centers to shaping their legal and regulatory frameworks, emerging markets have a real opportunity to drive future prosperity, believes Alathel.
Recent changes in Saudi Arabia stand as an example of what intentional redesign can achieve. Women’s roles in business and society have shifted significantly in a relatively short period. The critical lessons, she argues, are not cultural but structural. “Where systems are intentionally redesigned, women advance faster. Where legacy structures remain untouched, progress stalls.”
Forging future leaders
For young leaders navigating today’s fast-changing landscape, Alathel’s advice is uncompromising. “Do not optimize for comfort,” she says. “Seek environments that expose you to complexity early. Learn how systems fail, not just how they succeed.”
Employers must recognize changing leadership requirements. “It’s not just being tech-savvy. It’s about having empathy, being agile, being a risk-taker, being responsible.”
She advocates shadowing programs that give young professionals exposure to high-stakes situations without putting mission-critical projects entirely in inexperienced hands. “It allows you to observe how they operate under pressure,” she explains. “That’s how you discover talent.”
In an era when old boundaries are fast disappearing and old certainties no longer hold true, the ability to lead through risk is not optional. Neither is responsibility. As organizations and societies grapple with AI’s transformative potential, Alathel’s message is clear: leadership is not defined by control, but by accountability.
“Leadership,” she says, “is forged through responsibility.”
Patrick Woodman is editor of Dialogue
