Across industries, businesses are laying out plans to train employees to use generative AI and AI tools effectively.
Mercedes-Benz intends to invest more than $2.2 billion by 2030 to train employees in AI and grow data skills. McKinsey and PwC are upskilling employees to become better AI prompters and spot hallucinations. Commercial real estate firm JLL rolled out an internal generative AI tool for employees with accompanying demos and instructional videos.
But it’s not just individual contributors within an organization that need assistance. Knowledge workers, employees with technical expertise and high-level executives alike can benefit from support around using the technology as it becomes more ingrained in workflows.
More than 9 in 10 IT leaders and other upper management respondents currently use AI on the job, according to a Freshworks survey. Top tech leaders outpace their lower-ranking peers when it comes to the adoption of AI.
In response to mounting interest, vendors have released instructional videos and courses targeting skill gaps. In July, AWS launched a generative AI primer course for executives that comprised five videos covering foundational elements, historical context and use cases of the technology.
Whether an executive is speaking with colleagues about the technology or using generative AI tools themselves, it helps to have a baseline of knowledge.
Here are 3 key takeaways from AWS' generative AI primer course:
1. Transparency is key to AI implementation
While executives often helm strategies and sit with a bird’s eye view, generative AI experiments and implementation plans will require them to have boots on the ground, interacting with employees.
“The first step is communicating with your teams,” AWS said in the course. More than half of workers have “no idea” how their employer is using AI, according to a UKG survey of more than 4,000 employees.
“People tend to fear the unknown, so sharing your goals and how you plan to achieve them can go a long way to alleviating their fears and earning their trust,” AWS said.
AI transparency can improve workplace culture, with more than half of employees surveyed saying they would be happier and go above and beyond at their jobs, according to the UKG report. Nearly two-thirds said it would increase job engagement and satisfaction.
“Explain what’s in it for them,” AWS said. “Generative AI, like the cloud, is an opportunity for employees to transform their careers.”
2. Hardware changes fuel AI adoption
Hyperscalers are boosting their capabilities and infrastructure to support compute-hungry generative AI workloads.
While data centers have traditionally relied on central processing units to perform general-purpose computations, generative AI models require graphics processing units or tensor processing units. GPUs, initially designed to accelerate computer graphics, have made it possible to train models with billions of parameters due to their unique ability to efficiently run parallel workloads, according to the AWS course.
The focus on optimized hardware has positioned Nvidia as a leader with enterprise technology vendors leaning on the company for its chips. Vendors are also releasing their own versions of next-generation chip technology, such as AWS’ Inferentia2 and Trainium and Google’s Cloud TPU v5e chips.
“Specialized learning chips, which are purpose-built for high-performance, deep learning training have become far more cost-effective and accessible,” AWS said in the video.
Advancements in computing, coupled with access to big data and ML-optimized Transformer architecture, allows foundation models to be fine-tuned for a wide range of tasks.
3. Upskilling and guidance drive efficiency
Generative AI’s momentum has cast uncertainty over the future of some roles. However, vendors and analysts have underscored the use of these technologies as tools rather than replacements for human workers.
“Your people are the key to your success with generative AI,” AWS said. “Your existing employees are already familiar with your company, processes, people and systems.”
Providing employees with the skills needed to review generated responses for accuracy and bias while also providing them resources to become better prompters is the best way to bridge skills gaps, AWS said.
Technically skilled employees need training on emerging security risks and strategies for defending against generative AI-equipped threat actors.
Organizations can use generative AI policies to guide acceptable use. When Procter & Gamble rolled out an internal generative AI tool, it required employees to complete a mandatory 10-minute training session followed by signing an acceptable use policy.
