Exceptional EA

19th June, 2025

Mid-’25 reality check amid AI CEO’s warning about white collar careers

Today, I’m encouraging you to keep a cool head and appreciate AI’s positive impacts even as I bring you recent AI insights – some of them daunting – from a range of sources and topics, including the following.

  • an AI CEO’s prediction about the speed and intensity with which AI may impact white collar careers
  • Nobel laureate and “Godfather of AI” Geoffrey Hinton’s take on the potential extent of AI’s impacts
  • the recently published results of an international survey about AI-related employee attrition
  • the 2025 World Economic Forum (WEF) Future of Jobs report
  • the European Union’s EU AI Act (2023), New Nordics AI, and more

Effective risk management practices include scrutinising opportunities as well as risks, determining how much risk you’re willing to take (risk appetite), identifying and taking steps to mitigate risks … and repeating the process on a routine basis.

Just as blindly incorporating new resources can be risky, there’s risk in ignoring AI resources available to us. Whether or not you’re in a work environment currently making use of ChatGPT, Claude, CoPilot, Gemini, Grok and other AI tools, it makes sense to learn how and why to use such resources.

It also makes sense to continually scan the horizon and understand what may (or may not) be coming our way as AI continues to become increasingly intertwined with our careers and lives.

In Europe, where entities such as AI Sweden have been functioning since 2019, Fall 2025 will see the launch of New Nordics AI, an entity reflecting collaboration between leading organisations from Denmark, Finland, Iceland, Norway and Sweden. The mission is to accelerate AI adoption and innovation across the region.

It’s been suggested some countries are delaying regulation in order to boost competitiveness with China. The cynical among us might suggest the absence of regulations serves the interests of AI companies. While the European Union established the EU AI Act in 2023, most of our countries are still in the wild west when it comes to regulations. The EU AI Act represents the first comprehensive AI law anywhere around the globe and yet, despite armed conflicts on its (and other) doorsteps, this legislation specifically excludes military AI from its scope.

In the US, there’s a proposed Act containing a provision that would, as reported by Forbes, result in a .“…. 10-year moratorium on state and local regulation of artificial intelligence systems … that “… would prohibit cities and states from setting their own rules on automated decision-making, algorithmic bias, facial recognition or data privacy in AI applications.

I’m in the process of updating my cyber awareness presentation for an upcoming event, and am learning AI is increasingly factoring into cybersecurity operations. At the same time, many would argue AI is right up there with (and, some say, exceeding the threat of) employees as the most significant threats to cybersecurity.

When it comes to career risks, being uninformed is a big one. If you’ve seen me speak at a conference or in your workplace, you may have heard me speak about something I call informed optimism. Positive by nature, I’m also pragmatic. I believe that, the more informed and aware we are of what’s going on in the world around us, the better positioned we are to adapt and prepare for challenges and opportunities that may be coming our way. This, in turn, helps support both personal and career resilience.

Dario Amodei, CEO of Anthropic, one of the top AI labs globally, recently issued a warning even while touting Anthropic products. Anthropic is the entity behind AI resource Claude, which I’ve mentioned in other articles. One of the resulting headlines reads, “Behind the Curtain: A white-collar bloodbath”. I’ve read articles dismissing Armodei’s projection as self-serving PR for his company, yet I’ve also been paying attention to other sources and their perspectives on AI risks and opportunities.

As reported by CNN, Armodei projected that AI tools “… could eliminate half of entry-level, white-collar jobs and spike unemployment to as much as 20% in the next one to five years”.

Armodei was focused on entry-level positions rather than roles such as yours. It would seem, though, that a 20% unemployment rate would impact more than solely entry-level roles – and others have cautioned that AI will impact careers across org charts. You can find more of Armodei’s comments on Axios and, as always, I encourage you to exercise independent thought as you scrutinise the various sources I’m referencing here.

Nor is Armodei alone in projecting job losses within white collar careers. Expanding on its Future of Jobs 2025 report, the World Economic Forum (WEF) stated on January 7, 2025 that administrative assistants’ roles “… remain among the fastest declining … as generative AI rapidly reshapes the labour market.” Administration managers’ roles ranked 14th on the WEF’s list of “top largest declining jobs”, while the administrative assistant role landed in the number two position on the same list.

Yes. Amodei and his sister Daniela Amodei – the Anthropic president – are both former OpenAI VPs. It’s OpenAI that brings us ChatGPT, and the Amodeis are two of a handful of OpenAI employees who left that firm over “...differences over the group’s direction after it took a landmark $1 billion investment from Microsoft in 2019“. They co-founded Anthropic in 2021, and the firm’s governance statements include the following: “Anthropic is a Public Benefit Corporation, whose purpose is the responsible development and maintenance of advanced AI for the long-term benefit of humanity.” 

Speaking during the June 11, 2025 VivaTech tech trade show in Paris about Amodei’s comments, Nvidia CEO Jensen Huang expressed disagreement “… with almost everything” Anthropic’s Amodei said. According to the chip maker’s Huang, “Everybody’s jobs will be changed. Some jobs will be obsolete, but many jobs are going to be created”.

Geoffrey Hinton on job security: “Train to be a plumber”

I’m registered for and looking forward to a Frontiers of AI presentation next week by Geoffrey Hinton, the British-Canadian Nobel Prize winner who’s referred to as the Godfather of AI. Hinton is a computer scientist, cognitive scientist and cognitive psychologist who earned his PhD in Artificial Intelligence at the University of Edinburgh. A Professor Emeritus at Canada’s University of Toronto (U of T). Hinton is expected to speak next week to both the risks of AI and reasons for optimism.

For a decade, Hinton combined his academic career at U of T with a significant role at Google. He left that role in 2023 so that, in his words, “I could talk freely at a conference” about how dangerous AI could be. At the same time, Hinton is unequivocal in his conviction that AI will revolutionise science and medicine for the better. During his June 16, 2025 interview with The Diary of A CEO podcast host Steven Bartlett, Hinton identified paralegals’ and customer service reps’ jobs as among those on the chopping block.

We’re told that, just as earlier industrial revolutions saw machinery take over jobs previously reliant on human muscles, we can anticipate AI taking over what Hinton has referred to as “mundane human intellectual labour”. How much of what you do in a given day would you consider mundane?

So says Hinton, and I have respect for his thoughtful and highly informed perspectives. There are some who suggest jobs will quietly shift and that employers will simply not re-hire humans when vacancies occur by (an employee’s) choice. While that may be the case in some circumstances, have a look at the following results of recent surveys by the World Economic Forum and on behalf of international recruitment firm Adecco Group.

That’s according to the World Economic Forum (WEF) , which published The Future of Jobs Report 2025 on January 7th this year. The WEF found that, while 77% of employers plan to upskill workers in response to AI’s impact on business models, “… 41% plan to reduce their workforce as AI automates certain tasks.”

I found it interesting to read that 50% of employers expect to “transition” employees whose roles are exposed to AI disruption “… into other parts of their business, an opportunity to alleviate skills shortages while reducing the human cost of technological transformation.”

Hinton, in his June 2025 podcast interview, pointed to a past example of this approach. While we treat access to automated teller machines (ATMs) as a given, it was in 1967 that they were first launched in London, England. Until then, people had to do their banking during business hours, and lined up in order to do so with a human being. Hinton pointed out that this didn’t automatically displace the tellers, who went on to do interesting things.

When I walk into our bank these days, the ratio of ATMs to tellers/frontline customer service representatives sends a clear signal encouraging clients to use the ATMs.

While the WEF’s finding that 50% of employers expect to transition impacted employees sounds promising, take a bit of time to think about your workplace, and what other opportunities there might be for colleagues – or yourself – in situations where AI can and, indeed, may preclude the need for a human in the role.

Recruitment firm Adecco Group, which reports it has more than a hundred thousand clients and is involved in millions of careers each year, has reported more daunting figures. For a second consecutive year, the firm conducted an international survey involving 2,000 C-Suite leaders across 17 industries in 13 countries. In 2024, Adecco reported that 40% of workers were worried about long-term job security.

Flash forward to June 10, 2025, when Adecco reported that not only had 54% of leaders participating in its 2025 survey said their companies would employ fewer people within five years’ time because of AI; it’s already happening.

The same June 2025 Adecco release stated that 46% of the surveyed companies have already begun trimming employee counts.

Times change, yet some things remain constant. As we headed into the 2020s, it seemed sage advice to recommend coding as a great educational and career choice for someone graduating from high school. That’s no longer the case, as AI has already advanced to the point that Microsoft’s Copilot can create code for individuals as well as businesses.

Microsoft CEO Satya Nadella said this spring that engineers are relying on AI to write anywhere between 20 – 30% of the corporation’s coding. Meta’s Mark Zuckerberg has said his company plans to use AI for 50% of its software development in the year ahead.

There’s much in life that’s beyond our control. The great news, whether you’re job hunting or employed – bet it in an environment that’s bypassing AI advancements for now, or one that’s increasingly incorporating AI into its operating environment – is that you have capacity to hone and deploy skills employers have identified as core.

Returning to the World Economic Forum’s Future of Jobs Report 2025, you may find it helpful to know good old fashioned human analytical thinking is still seen as a core skill; it ranks first on employers’ list of core 2025 skills.

Second on the list of core skills employers prioritize in 2025 comes the cluster of resilience, flexibility and agility. You’ll find leadership and social influence rank third on the list of core skills, followed by creative thinking in fourth place, with motivation and self-awareness rounding out the top five. It remains crucial to maintain technological literacy, which ranks sixth this year, and empathy and active listening ranked seventh on the list. Curiosity and lifelong learning landed in the eighth spot among the top 26 core skills employers identified.

When you think about resilience, flexibility, agility, social influence, self-motivation and self-awareness, these are all skills I teach. Over the space of my almost three decades working primarily in C-Suites and – for the closing decade of that career – with a board of directors, I built and continually honed those skills.

Have a look at the training solutions I deliver, and explore the courses I’ve been providing the 2024-25 EA Accelerator (EAA) and EA Governance Accelerator (EAGA) cohorts. These cohorts are now closed, yet you’ll see I offer skills development training not only on minutes and other traditional aspects of the EA role; I also work with assistants who choose to develop expertise in governance and in all those highly ranked core skills identified by the World Economic Forum.

Are you looking for inspiring and practical professional development for yourself or your group? Check what clients around the globe have to say about learning with me. Then, click here to contact me, and let’s talk.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from EXCEPTIONAL EA: home

Subscribe now to keep reading and get access to the full archive.

Continue reading