Canadian Businesses Need an AI Blueprint to Increase Adoption

If 2023 is remembered as the year that AI went mainstream in the consumer market, 2024 will be the year that businesses got on this trend en masse.  

According to a recent report, AI is expected to spark a massive boost in productivity that will add up to  $16 trillion to the global economy. That number is likely much higher when you consider the business workflows that can be simplified and automated by AI, which is used everywhere to reduce costs, enrich customer and employee experiences, increase win rates, optimize supply chain performance, and more.  

“AI is expected to spark a massive boost in productivity that will add up to  $16 trillion to the global economy.”

Where does Canada stand in this scenario? Our AI journey is comparable to that of building a house. We have the right resources lined up — talent, leading researchers and academics, investment by both global and Canadian organizations, and government commitment — but we’re missing one critical piece:  our blueprint for AI adoption in business. 

Today, more than ever, we need a thoughtful plan to build up the strategic and trusted adoption of AI in business to ultimately deliver productivity and stronger performance. 

It’s Time to Build an AI Blueprint for Canadian Business 

Canada ranks in the top five globally in AI research and development, yet we lag when it comes to AI  adoption. According to a KPMG survey, overall adoption of AI among Canadian organizations is less than half of that in the US. Further, a recent study by Gartner, showed that only 54% of AI projects transition from pilot to production.  

“AI and automation can help reduce incident response times from days or hours to minutes, closing the gap between attack and response.”

To overcome these challenges, we must shift many of the processes and models we use in business including IT architecture, data management, and culture. AI presents the opportunity to make these  shifts, offering several opportunities for more efficient operations including: 

1. Creating exceptional customer care: AI can empower customers to get what they need quickly and through self-service, freeing employees to handle more complex requests.  
2. Processing claims with accuracy, speed, and efficiency: AI fine-tuned on business data can help reduce the time and cost of claims processing by automating and simplifying tedious and time-consuming manual tasks. 
3. Augmenting security teams to help detect and eliminate threats: AI and automation can help reduce incident response times from days or hours to minutes, closing the gap between attack and response. It can also help verify users’ access, find exposed assets, and enforce compliance measures. When teamed with a hybrid cloud architecture, it can help determine who gets access to what and when.

The companies that adopt an AI-first mentality in how they train their people and put things into production will outperform their competitors, now and in the future.  

Four Elements to Include in Your AI Blueprint for Businesses  

1. Invest in foundation models to be an AI value creator for your business

Any enterprise that wants to get the most out of AI should participate in the full value creation opportunity of foundation models rather than outsourcing their capacity, their strategy, and — most importantly — their data to third parties. Foundation models are, at their core, a new representation of data. This new representation has the power to unveil valuable insights from the data. Historically, the invention of new data representations has proved foundational to the advancement of science and technology, and the creation of new businesses. This is no different. 

“A business capable of creating its own AI models (and remember, they don’t all have to be huge) is a business that controls its destiny.”

Enterprises must carefully consider the advantage they could concede by allowing their data to be encoded in a foundation model that is not their own. Insights lie in that data. A business capable of creating its own AI models (and remember, they don’t all have to be huge) is a business that controls its destiny.  

Businesses train, tune, and govern their AI to consistently get the most out of these evolving technologies. In addition, as value creators, they have real ownership over the protection, control,  innovation, and monetization of what will become one of their most prized assets: enterprise foundation models encoding their most valuable data. 

2. Reimagine how work gets done in your organization

In Canada, we continue to face low unemployment as well as a shortage of skilled tech talent. Given this environment, it is critical we pause and fundamentally reimagine how work gets done. Canadian businesses need to look at this technology to see what opportunities it can create and what problems it can solve. 

For example, customer care teams can automate hundreds of thousands of call center responses with AI  with more than 90% accuracy and greater levels of customer satisfaction. HR teams could save thousands of hours each quarter by augmenting and automating what used to be labour-intensive data entry tasks. 

Cybersecurity teams will also have the ability to analyze data and defend against cyber criminals in real time as AI-enabled systems can detect fraudulent transactions as they occur, freeing up time for employees to tackle more complex challenges. 

3. Invest in the skills required to understand and benefit from AI 

Canada is a world-leading AI research hub brimming with opportunity. In the AI space, more than 35,000 new and innovative jobs are expected to be created over the next five years. While this is an exciting proposition, we continue to face a tech skills shortage. 

“Canadian executives estimate that 42% of their workforce will need to reskill as a result of implementing AI and automation over the next three years.”

This is not only a challenge in the talent market, but also internally at many Canadian organizations.  According to a recent IBM Institute for Business Value survey, Canadian executives estimate that 42% of their workforce will need to reskill as a result of implementing AI and automation over the next three years. Canadian executives surveyed ranked technology illiteracy as a top talent issue and confirmed that building new skills for existing talent is an important challenge for Canadian businesses.  

Many leaders have started to address this challenge by first evaluating how to increase technical skills in their workforce around AI. This includes ensuring a basic understanding of AI and its capabilities,  allowing them to work creatively and responsibly as critical thinkers and users of the technology.  

A mindset shift is also needed. AI is not meant to replace all work but to augment work. Ask team members to help identify tasks where AI or automation can help free up their time for higher-value tasks.  

We’ve started this exercise at IBM. Earlier this year, IBM’s HR team re-examined the highly manual and data-intensive quarterly promotions process, applying a custom AI solution to automate data gathering and thereby empowering human staff to devote more time to high-value tasks. 

4. Ensure the technology is trustworthy and responsible  

If AI is not implemented with care, the technology can reinforce biases, compromise customer trust and privacy, spread disinformation, and harm brand reputation.  

The backbone of any new technology is trust. To achieve that, any transformative technology must be brought into the world responsibly. For AI, this means careful implementations focused on specific use cases with proper controls and governance in place. Ethical considerations must be addressed from the outset. 

“We must be clear about how our AI systems are trained, what data are used in that training, and what goes into an AI system’s recommendations.”

As Canadian business leaders, we must deliver the highest standards of security, privacy, data protection, and compliance that are expected by our customers and government. We must be clear about how our AI systems are trained, what data are used in that training, and what goes into an AI system’s recommendations. Developing responsible and ethical AI requires that we remove any potential for human bias.  

Therefore, a critical step for any organization with a desire to apply an AI model is to establish guardrails to ensure that AI is transparent, governed, and trusted.  

Let’s Build a House to Last and an AI Blueprint for the Future

There is no doubt that we have arrived at an inflection point for AI — and the instinct of executives to act boldly is a good one. No one wants to be left behind by the march of these technologies. No business can afford to let this moment pass them by. But we must move forward and build this house on a solid foundation of trust, transparency, and empowerment through the implementation of an AI blueprint.