Artificial intelligence is rapidly transforming the way industries operate. Because of the speed of AI adoption and the technology’s ability to alter human and capital investment within a business, AI forces leaders to make important decisions about where the people on their teams drive value.
Transitioning from traditional human-intelligence-driven methodologies can cause trepidation even among the most adventurous teams. Of course, like any new technology, there are always the risks that come with budget, deployment, integration and QA/QC to contend with. To prepare your organization for the coming changes, you must communicate your strategy and have pilot projects on deck that allow your teams to become familiar with the technology. These components are the keys to building momentum and trust in the AI system.
We saw AI move into the language industry about 2-3 years ago, and adoption has accelerated in a way few technologies have. It’s now used on more than 50 percent of translation projects, directly replacing the traditionally human work of translation. Ultimately, the industry has learned that AI provides significant benefits, but it hasn’t been an easy, or painless, path.
Putting AI to work is complicated by the fact that technology is constantly evolving and that its adoption speed is much faster than previous technology. Just as our human teams get close to “normalizing” an AI workflow, the technology changes. Additionally, for AI to be “intelligent,” the system needs to be trained, which means you need a strategy for data retention, security and application.
Finally, because it’s relatively new technology, finding and retaining the right team isn’t easy. Consider creating a dedicated AI team composed of internal stakeholders at multiple levels of your business and possibly even an outside consultant to act as a guide and anchor point for the broader business. This “expert” team can help build the confidence your people have in AI and your AI strategy.
AI Is Changing the Way Industries Operate
To be successful with AI, we first have to shift the way we approach innovation. With the speed of change and adoption rates within the industry, we must adopt a “fail fast” investment model of testing and learning from failure—tweaking, scrapping or starting over as necessary.
Due to the rapid pace of its technological evolution, a slow, methodical implementation strategy is often out of date by the time we get to deployment. It impedes iteration and restricts the refinement of products and services, creating a negative impact on our competitive market position. Failing fast teaches us to be responsive to shifts in technology and actively adopt them rather than resist them as we deploy new workflows. It makes our system and our teams more resilient.
We also find it helpful to prepare teams for discussion about how AI affects human capital needs and how AI will change the way people work. Because we lacked experience in the beginning, we tended to take a more traditional approach of “we learn, then teach.” But we needed to adopt a process of all learning together and using the “teachers” to act as central hubs of information. Rather than coming in and saying, “Do X or Y,” consider, “We have seen X or Y work well for this group. Could it work for you, and can you find anything we can share with those teams?”
Centralizing AI knowledge through a core team is profoundly useful, and we will use this method for other solutions, but it was a real shift in approach for our experts. Bringing AI to teams through mentors helped us maximize the technology and respond to new features and functions far more rapidly.
Perhaps because of our initial top-down approach, we did get a lot of resistance and a decrease in engagement at first. People actively didn’t like AI solutions, even before they had seen them; there was a perceived threat it would take away their jobs. There is truth to that; AI is likely going to change a lot of jobs and eliminate many traditional roles. It’s important to address this within your strategy, and this is an area where peer discussion and central information teams helped.
We looked at what AI could do and where it had challenges, mapped roles and functions to the areas the AI struggled, and produced a plan to migrate resources to these more complex tasks. Training employees for these tasks provides a higher value to the company, customers and employees.
We also found that separating AI tasks from human tasks lowered the costs (though not profitability) of AI tasks while increasing the profitability of human tasks. Ultimately, this increase in profit can fund and support better careers, working standards and training options for the company if they are willing to invest in their employees.
As an example of what an AI workflow can look like, an international energy company with 58,000 employees across the globe was litigating a case with a team of 100 lawyers without a central translation solution. More than 10 languages and millions of files were involved, leading to duplicate translation requests and more than 13 million words to translate, which would have taken humans multiple years to complete.
ULG created an AI-driven digital library for translation requests, trained the AI specifically on key legal phrases within the content pool, and used data from 15 years of legal translation to ensure the AI could act intelligently with this specific type of content. The system could track files, extract content, identify duplicates and auto-translate every word. Files could be sent to a human translator for traditional processing or review and confirmation. Given the volume and timing, we still needed more than 20 linguists to do system training, review and confirmation. The result of the overall AI and human-powered solution was that the legal company easily met its production time frames, provided a quality the court could accept and it experienced significant cost savings.
Set Yourself and Your Company Up for Success
What problems will AI solve for your business? Auditing where human intelligence is being inefficiently used or wasted—or even where AI could fill gaps before implementation—is a critical first step. You need to have a clear understanding of how AI will work for and with your teams.
By identifying clear opportunities for success, you define how AI is a vehicle for creating change for your people, improving workflows and creating a culture of possibilities for their professional development. AI involves highly complicated decision-making, and it presents an opportunity for the company’s leadership to leverage the skill sets of your department heads and their people.
Your team’s deep knowledge can be used for optimal AI performance by employing strategic human intervention, and AI can introduce intellectually challenging and engaging work that serves the individual’s interests and yours. Every group in your organization almost certainly has a list of inefficiencies they want to solve, and AI is the most likely answer. It can free them to concentrate on more mission-critical tasks and empower them to focus on outcomes and solutions.
AI is inevitable and is happening quickly as all companies aggressively seek competitive advantages. If you fail to implement an AI solution, you risk falling behind industry trends. Finding the true balance between the value of human intelligence and AI has a profound effect on the business by driving profitability, the speed of customer responsiveness, and the creation of new services and solutions for your customers.