Professionals today face a flurry of AI tools, platforms and features, each promising to transform the way we work. But not every one of those innovations delivers. Many organizations fall into the trap of investing in trendy AI solutions that fail to meet their needs, ultimately draining time, money and energy.
“People are often looking for the best model or the newest tool, but what matters more is that it makes sense for what you’re trying to do,” explains AI strategist Parnaz Tabrizian.
Tabrizian, who serves as Director of Innovation Labs at Alberta Investment Management Corporation (AIMCo), has helped teams across sectors build, evaluate and implement AI tools. At Concordia Continuing Education, she brings that deep expertise to the classroom to support professionals in building clarity and confidence in AI adoption.
For Tabrizian, adopting AI isn’t about grabbing the latest release. It’s about the discipline of forecasting value by understanding which capabilities are becoming essential in a given field, why they matter, and where they can meaningfully improve your process today.
She says the difference between trend-chasing and true productivity comes down to clarity of purpose. She’s seen teams pour resources into developing custom tools, only to have a cheaper off-the-shelf solution solve the problem a month later.
She’s also seen companies rush to adopt a trending platform that adds friction, not value, because it doesn’t align with how the work actually gets done.
That misalignment, Tabrizian argues, is the pitfall many organizations are stumbling over. Not missing out on AI’s full potential, but simply buying in based on misguided reasoning.
She offers a pragmatic starting point.
“Look at what you do every day,” she says. “What are your repetitive tasks? What takes time that doesn’t need judgment? That’s where AI automation makes the most sense.”
From there, it’s about experimentation. Tools like chatbots, Tabrizian says, require constant refinement.
Parnaz Tabrizian, AI strategist
“When people use tools like ChatGPT, they get frustrated if the first answer isn’t good,” she notes. “But that’s normal. You give feedback. You try again. Eventually you know how to instruct it, and you get real efficiency.”
That principle applies at every scale. Whether it’s writing an email or integrating an AI agent into a workflow, it’s not the tool that determines success — it’s the feedback loop.
Is the tool learnable? Can the user refine the output over time? Does it support better thinking, or just more output?
She says that one of the most effective ways to forecast value is to map the current process before introducing AI. This helps teams see what can be automated and what shouldn’t.
“If you don’t truly understand how the work gets done today, AI just adds noise,” she says.
One of the most common errors Tabrizian sees with organizations and teams is them mistaking speed for strategy. She points to marketers and communications teams who use AI to generate more content, more quickly. Later they realize it doesn’t connect to business goals.
“That’s just being faster,” she notes. “It’s not the same as getting better.”
AI tools are often positioned as efficient assistants. But Tabrizian advises treating them like junior employees. They can be quick and helpful, but they need supervision.
“You can’t hand over a process and walk away,” she said. “You still need to be the one making sure the result is useful, ethical, and accurate.”
When assessing tools, Tabrizian looks at three things: relevance to the task, potential for improvement and ease of integration. A tool that saves five minutes but requires an hour of troubleshooting is a bad bet. But a tool that gets 80 per cent of the way there — consistently — is one she’ll build around.
That thoughtful, iterative approach to forecasting the potential for AI is what separates successful integration from tech churn. AI isn’t going to stop evolving, she says. But the people who get value from it aren’t the ones chasing features. They’re the ones asking better questions.
“The difference between someone using AI well and someone posting about AI on LinkedIn is just that one of them wrote about it first,” she said. “You don’t need to know everything. You just need to be trying things that make sense in your context.”
© Concordia University
