AI … WOLF IN SHEEP’S CLOTHING?
To augment or replace, that is the question!
“AI won’t replace people—but people who use AI will replace people who don’t”.
This statement in a recent IBM report ‘Work for an automated, AI-driven world’[i] holds out the hope that as long as we can use AI (or we can learn to use AI) then there will be a role for us in the future workplace.
Recent Goldman Sachs research suggested [ii] generative AI has the potential to bring about sweeping changes to the global economy, driving a 7% (or almost $7 trillion) increase in global GDP and lift productivity growth by 1.5% over a 10-year period. So the mantra of positivity goes, enlightened organizations will see a rise in productivity through a work enhancing human-AI symbiosis.
But wait a minute … at the risk of being an ‘AI Cassandra’ we do need to consider a potential alternative reading of the runes! Consider for a minute a recent study [iii] from Stanford’s Human-Centered Artificial Intelligence group (‘HAI’) and the Massachusetts Institute of Technology. They studied the impact of generative AI in a customer service setting, reporting that access to AI assistance increased call center agent productivity by 14%. So much for Goldmans 7%! But dig a little deeper and these gains mostly accrued to novice or less able workers. The research found few positive effects of AI for the highest skilled or most experienced members of the company.
“This may be because the AI model disseminates the potentially tacit knowledge of more able workers and helps new workers move up the experience curve.” Lindsey Raymond, Co-author
“High-skilled workers may have less to gain from AI assistance precisely because AI recommendations capture the knowledge embodied in their own behaviors” Erik Brynjolfsson, Co-author
In his 2018 paper ‘Artificial intelligence and the future of work: Human-AI symbiosis in organizational decision making’[iv], Cornell Associate Professor Mohammad Hossein Jarrahi argues that the short-term efficiencies offered by task automation could lead to long-term unintended consequences such as the loss of valuable expertise. But what the HAI/MIT research perhaps shows is that much of this valuable expertise/experience is no longer as valuable as it was. AI is benefiting less experienced and less skilled staff - so they get a boost to their productivity and in parallel it is marginalizing skilled workers who may find that this AI driven ‘levelling up’ is taking away some of their former comparative advantage.
Does that matter? Well, we may assume these more skilled, more experienced staff are correspondingly more expensive. What do we think is likely to happen to them … now that someone new in the door augmented by AI can do the job just as well? The question we have to ask ourselves is simply … as AI does more of the heavy lifting, do we believe that organizations will resist the temptation to reduce headcount and/or reduce the cost of labor?
If we are looking for cues to help answer this question, we might look no further than BT’s announcement in May 2023 that it was planning to cut as many as 55,000 employees by 2030, with 10,000 of those being the result of the adoption of AI.
“By the end of the 2020s BT Group will rely on a much smaller workforce and a significantly reduced cost base […] the new BT Group will be a leaner business” Philip Jansen, CEO
What starts with call center staff today, may soon spread to many white collar activities … accountants, lawyers, copywriters etc.
And so AI may well augment employees, but which employees? The more experienced, more skilled (in the ‘old world’), the more expensive? And will we stop with augmentation? There is another quote that we have seen attributed to Diane Yoon at OpenAI (the creators of ChatGPT) … “Your job will be lost to someone who knows how to use AI … before it is lost to AI”
Keynes introduced the concept of technological unemployment which is unemployment stemming from our discovery of the means of economizing on the use of labor outrunning the pace at which we can find new uses for labor. This was just an echo of the same theme set out by David Ricardo 120 or so years earlier.
As Erik Brynjolfsson said in the HAI/MIT paper… “We don’t know if the impact of AI on productivity may vary over time, and adding these tools to the office could require complementary organizational investments, skills development, and business process redesign. There’s so much we don’t know.”
To contribute in a small way to the understanding of where businesses sit vis-à-vis skill development to better capture AI derived opportunities, the authors have undertaken a small pan-European research study of their own[1] to better understand to what extent businesses have started to upskill their employees. A few headlines …
Fewer than 1 in 5 respondents believe AI will radically change their business. 40% see it having a ‘tactical’ impact at most.
Fewer than 1 in 5 respondents believe their business is approaching AI as a top priority
Fewer than 1 in 5 say their business is actively upskilling or reskilling the workforce – almost 50% are currently ‘sitting on their hands’
Does this data suggest that 20% of businesses are early adopters … seeing AI as having a radical impact and doing something about it? Well, no. Drill down a bit further and we find that less than half of businesses in which AI will drive “radical change” see “investing time and resources into incorporating AI” as a top priority and less than half of these businesses in which AI will drive “radical change” are proactively upskilling and reskilling employees. This may suggest that for many businesses the radical change that they envisage does not have employee skills at its core?
On an individual level … 15% of respondents believed AI would totally redefine their role/skills – but of these only 55% were being proactively upskilled. Of the 28% of respondents who believed AI would have a “major impact on virtually every aspect of their role”, only 19% were proactively upskilling or reskilling.
People in higher managerial roles are more likely to see AI radically changing their business and redefining their role and the skills they need. But they typically see this impact as more likely to be positive for them personally. Respondents in C1 type roles (Supervisory or clerical/ junior managerial/ professional/ administrative) believe AI will have less of an impact on the business BUT they see the impact of AI being far less positive for them personally.
The data suggests that the arrival of AI is such a massive event that people are finding it difficult to understand its impact (it is outside their frame of reference) and so responses tend to fall into two camps
1. Let’s bury our head in the sand and we will worry about that tomorrow … taking a ‘play it by ear’ approach to tactical and strategic change.
2. Let’s take some limited action because we can see how an aspect of AI can comparatively easily be added to improve something we do e.g. add creative prompt engineering to help generate faster ad-copy
There is a lot of talk in academic literature and news articles about the future belonging to those who will ‘upskill‘. But what doesn’t seem to be happening is for people to engage with precisely/ exactly what this upskilling means. Few people are making the point that this is a massive shift in what is required of us humans. It isn’t just about sharpening up thinking skills and being a bit more creative and adding a little bit of edge to what we’re currently doing. We are now staring down the barrel of a fundamental overhaul of what constitutes valuable skills. This should worry not only businesses but also schools and universities … but we suspect for now this will be relegated the ‘too difficult box‘! Who will be brave enough to break cover and grasp the nettle?
The authors in a series of recent articles have been reimagining the business skills needed for growth. We believe that our human ‘super-power’ is the ability to comprehend - and importantly, empathize with - the complexity of the human condition and experience, conveying it in a way that businesses can understand and action. In turn, this ‘super-power’ draws upon 7 ‘Power Skills’.
These are the skills that we believe are most important to dial up because they are the ones that allow us to work most productively alongside generative AI. Maybe one day, even this reboot will be overrun by the power of AI, but at the moment, arguably, these are the areas we should be focusing on in working alongside generative AI to build growth.
Each of these skills has required a fundamental philosophical, methodological and pedological core reboot of what this topic is all about. Take for example something straightforward like the ability to undertake a piece of data analysis. We might have traditionally seen this through the lens of data analysis / statistical analysis / IT / data science etc. BUT it will actually require a fundamental rethink that takes into the world of ‘sensemaking’, ‘fuzzy logic’, abductive reasoning and the three arrows of time (see our earlier articles) i.e. an altogether totally different approach to making sense of vast amounts of imperfect information in the face of uncertainty and disruption.
In embracing the 7 Power Skills – this goes further than just ‘upskilling’.... it’s a radical rethink of how we get the most out of the creative human brain in the context of the support that it / we can now get from AI
[1] n=1100 respondents working in businesses with at least 25 employees. August 2023
[i] https://uk.newsroom.ibm.com/2023-08-14-IBM-Study-Drives-Massive-Shifts-in-Jobs-and-Skills-as-Employees-Priorities-Meaningful-Work#:~:text=The%20study%20%E2%80%9CAugmented%20work%20for,issue%2C%20according%20to%20surveyed%20executives. Survey of 3,000 global C-Suite executives across 20 industries and 28 countries
[ii] https://www.goldmansachs.com/intelligence/pages/generative-ai-could-raise-global-gdp-by-7-percent.html April 2023
[iii] GENERATIVE AI AT WORK Erik Brynjolfsson Danielle Li Lindsey R. Raymond
[iv] https://www.sciencedirect.com/science/article/abs/pii/S0007681318300387
 | A guest post by
|