Crossing the Bridge

The corporate world is experiencing an AI paradox. While almost every organization is tinkering with tools like ChatGPT, Claude, or Copilot, the vast majority of AI initiatives, (95%, as last summer’s well-known MIT paper estimated), are not actually paying off in terms of organizational performance. According to Walid Hejazi, Professor of Economic Analysis and Policy at the Rotman School of Management and lead faculty on many of Rotman’s open enrolment and custom AI programs, the problem is not the technology itself, but a failure of role redesign and a lack of personal mastery at the executive level.

The Productivity Gap

Hejazi floats a scenario of a company with 50,000 employees where a new AI tool saves each person a few hours a week. On paper, that is a massive gain. However, if roles are not redesigned to capture that saved time, the organization sees no bottom-line benefit. The employee simply does their old job faster or less stressfully, while the promise of AI remains unfulfilled on the other side of what Hejazi calls ‘The Bridge’.

The Bridge, crosses the gap between just using AI and truly capturing its value. According to Hejazi, to cross this bridge, executives need an organizational level strategy and a plan to encourage individual adoption that serves that strategy. First, they must understand it through the lens of their own personal productivity. When a leader understands how to use these tools for their own workflow, they gain the insight necessary to understand how their employees are using the tools and also why they might be resisting them. The fundamental insight “facing every professional today is not whether they will be replaced by AI, but whether they will be replaced by a person using AI more productively than them,” says Hejazi.

Avoiding ‘Copilot Purgatory’

One of the most significant obstacles on this bridge is what Hejazi terms ‘Copilot Purgatory’, which occurs when employees view AI-driven efficiency gains as a threat to their job security. If a manager talks about ‘productivity’ and ‘efficiency’ without sharing a clear plan for repurposing rather than replacing staff, employees may well quietly ensure the AI pilot fails to protect their roles.

There is also the rise of ‘Shadow AI’. Hejazi tells the story of a massive company that banned all employees below the VP level from using AI due to data privacy concerns. This is a ‘gate’ approach, where arbitrary rules dictate the use (and non-use) of AI in organizations. The reality? If you ask an employee to summarize a 100-page document, they are going to use AI to do it regardless of the ban.

The solution, Hejazi argues, is not ‘gates’ but ‘guardrails’. "Create guardrails and get out of the way" he suggests. By setting clear rules for data use and ethics, organizations allow for innovation while maintaining safety, and provide autonomous space for employees to give them meaning at work.

From Replacement to Augmentation

This shift in mindset was underscored during a recent high-profile dinner Hejazi attended with Nobel laureates, the physicist Geoffrey Hinton and economist, Daron Acemoglu. Acemoglu noted a concerning trend: the world's biggest AI companies are currently focused on replacing people.

However, the real opportunity needs to be on AI collaborating with employees, not replacing them. That is ‘augmented intelligence’. The emphasis needs to be on technology designed to make labour more productive rather than redundant. We need to remind ourselves that we are still in the very early days of dealing with these issues; as Hejazi quotes his colleagues “we are currently in the ‘in-between times’, a period similar to the 40-year lag it took for electricity to truly transform the economy after its invention”.

The Three-Hour Strategy

So, what does this look like in practice? Hejazi shares a striking example from a recent Rotman session. An executive was asked to bring a high-level strategic idea to the table. Using a series of five or six different large language models (LLMs) and highly developed prompting techniques, she transformed that raw idea into a polished PowerPoint presentation, complete with financial projections and a board-ready strategy, in just three hours.

This was not just about doing work faster; it was about better judgment. When the ‘drudge work’ of drafting and formatting is handled by the machine, the executive is free to iterate, look at the problem from six different angles, and find the ‘blind spots’ that a human-only process might miss.

It also changes the nature of value. Hejazi mentions a former student who built a successful business and wanted to decide whether to scale or sell. When hiring consultants, he told them: "Don't give me a standard consulting company PowerPoint, because I can do that with ChatGPT in two minutes". In this new world, you are no longer paid for the ‘commodity’ of information, but for the insight and judgment you layer on top of it.

Future-Proofing the Professional

For those looking to lead through this transition, the path forward involves leaning into the technology rather than waiting to see what happens next. Whether it is using AI to fact-check academic papers for plagiarism or using ‘prediction technology’ to prevent grocery store customers from weighing expensive grapes as cheaper bananas, the applications are as varied as the professionals using them.

Crossing the bridge to the AI era requires a foundation of personal mastery. By understanding the limitations and the potential of these models firsthand, leaders can move their organizations out of ‘Copilot purgatory’ and into a future where technology does not substitute for human talent, but dramatically complements it.

For those ready to build that foundation, Rotman has a number of AI programs for executives looking for a high-touch, practical environment to master these tools. It is an opportunity to move beyond tinkering and start building the guardrails for a more productive future. 

---------------

Explore more:

View all of Rotman's AI-related executive programs here

AI Powered Productivity for Professionals - a 1 day in-person workshop

Generative AI and Organizational Transformation - a 3 day in-person program