In the last 5 years, there have been significant advances in artificial intelligence (AI) — from Google DeepMind’s AlphaGo besting a human world champion Go player (significantly more complex and challenging for AI than IBM’s Deep Blue win over Gary Kasparov in 1997), to social robots like Jibo and online chatbots. With massive M&A and investment activity in AI-related companies, we are starting to see decades-old visions of this technology starting to take shape in tangible innovation throughout areas such as education, health care, manufacturing, and business.
Of course, with any new advancement of technology comes both correspondingly overhyped expectations on the one hand, and dire consequences of its societal impact on the other. The latter has even spurred prominent science and technology figures such as Stephen Hawkings, Elon Musk, and Bill Gates to sign an open letter advocating for responsible AI oversight in order to mitigate risks and ensure the "societal benefit" of the technology.
Their concerns can be read in detail in the related "Research Priorities" document, but the chief concern is the economic impact of AI. More specifically, understanding how AI will "affect the wages of less skilled workers, creatives, and different kinds of information workers".
While some may think it is premature to be concerned whether AI will actually replace information worker jobs, we are already starting to witness some initial implications of how AI will impact information technology across areas including…
- Analytics: using AI algorithms and predictive models to help sift through vast amounts of data to find the patterns, correlations and anomalies that help to surface insight and enable employees to take action.
- Security: using machine learning to identify potential threats and learn how to avoid false positives so that attacks may be better thwarted proactively and efficiently.
- Information Retrieval: using natural language processing to enable employees to verbally ask for operational data, such as my top 10 customers, or the most red accounts, and have a backend server fetch the relevant data from multiple IT systems and present it.
There is no doubt that AI-based solutions will have a significant impact in terms of tools that IT uses and their job functions, and will eventually start to impact employees. The transition, however, may take some time, and I believe will occur over three broad stages:
- Process impact: where the use of AI tools will become so prevalent in certain business processes, that job tasks will be altered to accommodate or reflect the use of these tools (such as optimizing resource allocation).
- Productivity impact: where the AI tools become so efficient at performing a subset of tasks that humans do, and thus result in less employees needed to accomplish the same thing (such as automated, model-based stock trading).
- Job replacement: the point at which AI tools completely take over a specific job function and perform it at a level where a human is no longer required (such as autonomous delivery vehicles)
It will be interesting to see how AI plays out and the impact it has on IT. I suspect that in the near-term, benefits of AI will outweigh the cons, with increased productivity and improvement in job satisfaction as software and applications become "smarter" and more helpful. In the longer term, there is certainly a risk of job loss from those tasks that can be automated with AI advancements. However, society has witnessed this type of transition in the past, as we moved through agrarian to industrial to service-based economies. It may be a difficult transition, but we will find a way to adapt, and there are still some jobs that AI cannot replace, for now.