AI in the Workplace
A recent article published in Harvard Business by Deborah Perry and Josh Drean discussed a series of forces that are changing how we work.[1]
Key Forces Shaping the Workplace
These forces are:
1) the acceleration of productivity through artificial intelligence (AI),
2) The introduction of Web3 business models,
3) An upcoming generation of workers who blur the lines between real and digital work, and,
4) A societal shift in how we perceive work.
The way that companies create wealth is rapidly changing with new business models driven by technological advancements and new paradigms. As a result, work is undergoing a profound and irreversible transformation. This article discusses the impact of AI in today’s work environment.
We will take the definition of AI as “the ability of a computer or a computer-controlled robot to perform tasks commonly associated with humans”.[2]
The Origins of AI in the Workplace
AI has made headlines recently, but its origins can be traced back decades or even centuries. Much of the effort to create technology with artificial intelligence has been based on the premise of simulating and enhancing human capabilities.
This takes us back to Aristotle, around 300 B.C., when he proposed deductive reasoning principles called syllogisms, which laid the foundation for the development of logic.
At the same time, linguistics addresses the most important structures of human language and aspects related to comprehension and communication.
Historical Milestones in AI
- 1930s: Alan Turing developed the theoretical concept of the Turing Machine, foundational for modern computers.
- 1950s: Turing devised the Turing Test, aimed at determining if a machine could imitate human responses convincingly.
Compare this progress to today’s chatbots and advanced language models like ChatGPT. The strides in AI are evident when interacting with these tools, showcasing its transformative potential.
I recommend reading the book The Innovators, a magnificent account of the development of computational technology, the internet, and the people who made it possible.
AI Today
It’s worth noting that artificial intelligence has been around for many years. Examples include:
- Assistants like Apple’s Siri, Amazon’s Alexa, or Google Assistant
- Tools like Google Translate or Netflix’s recommendation algorithms
Unlike other tools, ChatGPT lets users ask questions in natural, conversational language. Even more surprising is that its responses feel human, leading some to wonder if these developments can pass the Turing Test. Developed by OpenAI, ChatGPT quickly disrupted many fields, not just business. Since then, new versions and alternatives from other companies have emerged. But, how does it affect my work?
Simply put, as a close alternative to imitating humans, the initial concern is that we could be replaced by technological alternatives–whether it’s software like ChatGPT or a combination of software and hardware, such as specialized machines and a sort of robots.
Remember that one of the previously mentioned premises is that AI can imitate and improve human capabilities. So, why hire an expensive human whose productivity today might be affected by their mood when I can rely on an AI instead? The possible answers raise fears, some of which are reasonable. This is not the first time this has happened.
How AI Impacts Work
Before the disruption of AI, there were many instances when technology presented changes in the workforce. Some of these changes were minor whereas others were revolutionary. Regardless, every time this happens, especially in a disruptive manner, the fear of negative impacts on the workforce arises. And rightly so.
During the Industrial Revolution, many jobs were eliminated due to the arrival of the steam engine. There were even moments of violence. For instance, back in 1812, the Luddites (a group named after their leader, Ned Ludd) opposed these changes. They damaged factories and machinery in protest against the impact this technology had on their employment sources.
Undoubtedly, many jobs disappeared during that time, especially for unskilled workers. On the other hand, new positions were created. The use of machines required new skills, so these positions were significantly better paid, for both women and men.
Delving deeper into the topic, not everything was rosy. Inequities persisted, especially in payments to women and child labor (remember, it wasn’t until the 20th century that laws regulating child labor were created). Still, more jobs were created, and, generally, they were better paid. However, this also meant that workers had to master a new set of skills that were not previously required.
More recently, technology has had profound impacts on the agricultural sector in developed economies. Take the case of the USA. In 1900, 41% of the workforce was employed in the agricultural sector. Yet, by 2000, only 1.9% were in this sector.
Now, in 1900, nearly 22 million working animals (mules and horses) were required. Compare this to 3 million in 1960, with the rest of the work compensated by 4.7 million tractors. And productivity increased as well. The US went from an agricultural export index of 20 in 1915 to nearly 160 (8 times more) in 2000.
The technification in the agricultural sector increased productivity and reduced the need for labor. Consequently, the economy created new sources of employment in other sectors, many of said jobs did not previously exist.
The Impact of AI
If the results in the past, although some were contradictory and had their bad moments, generated long-term progress and greater well-being for societies.
So, why is it different this time?
Basically, AI presents enormous potential to replace–and improve–tasks performed by people. Some experts talk about AI in stages: first weak AI, then strong AI.
Where “weak AI, also called narrow AI, is capable of performing a specific task that it’s designed to do. Strong AI, on the other hand, is capable of learning, thinking, and adapting like humans do. That said, strong AI systems don’t actually exist yet”[11].
In fact, experts talk about a third stage, also hypothetical today, called Artificial Superintelligence (ASI). It refers to a computer with cognitive functions and thinking abilities that are superior to those of a human being. But let’s take it easy; some believe that Skynet is about to become self-aware and John Connor needs to come to save us.
Although we must consider scenarios two and three, our immediate reality consists of weak AI applications. Platforms like ChatGPT, Perplexity AI, Gemini, and YouChat are a few of the most popular conversational AIs. Though advanced, these are still weak AI applications, providing only quick and often accurate answers to questions inputted as chat instructions.
As we will see, it is about understanding and leveraging new technologies rather than harboring fears or speculating about negative impacts. In fact, AI will indeed profoundly change the way we work, but changes remain speculative. There is no clarity or consensus on the impacts it will have.
There is talk of a myriad of jobs being replaced by AI. Similarly, there is word about a huge number of jobs that result from it, but the debate remains open. What is clear is that those who struggle to adapt to these new tools will be left behind and perhaps doomed to failure.
Current Impacts of Weak AI on Organizations
As established, current applications can only support a series of specific tasks for which they were designed. Even so, today’s AI has strong potential to generate positive impacts in activities such as:
1) automating and supporting repetitive tasks,
2) reducing errors,
3) data analysis, and
4) simplifying and improving processes, to name a few.
For example, a remote worker performing digital marketing activities for a client in the U.S. could benefit from these tools.
They could rely on AI to create personalized advertising campaigns, analyze performance metrics, and optimize social media content.
Challenges Related to AI
Continuing with the previous example, the impact and productivity improvement can be significant for both the company and the remote worker. However, there are also important challenges to capitalize on these benefits.
Some of these include:
1) The company’s openness to using AI tools (as well as to employing remote workers in this example); and
2) The worker’s openness to researching and training in the use of AI tools.
Before listing other challenges, it's important to emphasize that these first two are fundamental for leveraging AI, as with any other technology. A mental openness to new options, technologies, ideas, and perspectives is necessary to see the world in new ways. This involves accepting that we are living through rapidly changing times in nearly all areas of our lives.
Once you have the right mental framework for integrating AI into your work environment, consider the following additional challenges:
3) Lack of knowledge or understanding of the context.
Current AI models lack true contextual understanding, which can lead to misinterpretations or incorrect recommendations.
4) Dependence on training data. Models require data to be fed into them. That said, if the data is incomplete, erroneous, or biased, it can lead to flawed conclusions.
5) Ethical dilemmas. A broad and deep understanding of the subject matter, its context, and the ethical values and principles involved is required. While this may not be necessary for transactional issues, it is important for more complex problems.
Conclusion
AI is here to stay. It’s already transforming the workplace and will continue to do so. Understanding and adapting to AI is essential for organizations. Leaders must prioritize training and integrating AI tools to remain competitive.
Harvard professor Raffaella Sadun highlights the importance of distinguishing between AI’s actual and potential impacts, attracting the right talent, and balancing workload with upskilling.
How is your organization preparing for the challenges and opportunities brought by AI in the workplace?
