Dear employees, re: AI, it’s all going to be okay (in the end)
It’s about to get messy. Generative AI and agentic AI is going to have a significant impact on employee engagement, and the businesses that have laid the groundwork will survive, and help their people survive.
I was delighted to see IBM’s Institute for Business Value report, “From AI Projects to Profits”, last week. It’s extensive, and it explores the way that businesses are moving from the experimental phase to the core phase. This is where the rubber is going to hit the road and where we’ll see dramatic changes in the working landscape.
The natural reaction – from an employee perspective – is one of fear. “This is going to take my job!” will be the cry, and it will.
For us as internal communicators, employer brand managers, and HR & transformation directors it means that we need to be honest with our employees about how this will affect them and create the cultural bed for change.
In this blog, I’ll elaborate on the following six tips to creating an AI-ready culture:
- Be honest. Everything WILL change.
- Take control. Humans can revolutionise AI.
- Create urgency. AI needs our critical minds NOW!
- Stay informed. Sharpen up your data literacy and critical thinking!
- Channel emotion. Humans do empathy far better than AI.
- Learn to talk to AI. It’s all about language.
So, let’s get into it.
Everything WILL change
As Western Union said in a report:
predictions, while they sound rosy, are based on wild-eyed imagination and lack of understanding of the technical and economic facts of the situation, and a posture of ignoring the obvious limitations of (this) device, which is hardly more than a toy.
Then again, they were writing about the advent of the telephone in 1876. You see, we’re suspicious of change. Always have been.
But there’s no escaping the fact that roles will go. Jobs won’t, but they will change dramatically.
AI has already revolutionised my industry.
Research can now be done at 100 miles an hour. Themes can be summarised and distilled in seconds. Words can be written in pretty much any tone of voice and content can be created in the form of film, gifs, music (add as appropriate) at a dramatic pace.
In fact, we humans will need to leave a few misstakes in our work to highlight that it’s really us speaking. (I hope you see what I did there.)
But AI is not a threat! It’s an opportunity.
We can do so much more, if we know how to use it.
That’s because the true power of AI is unlocked only when paired with strong human skills. Especially critical thinking and effective decision-making.
As our Brand Activation director Sally Tarbit says, it’s the “powerful blend of divergent thinking (creative ideas) and convergent thinking (critical evaluation) creates unique, evidenced, and effective outcomes that ultimately lead to greater brand value.”
Humans can revolutionise AI
AI is revolutionising access to information, data analysis, and decision-making processes to enable faster, more accurate decisions, and that in itself causes problems.
The sheer volume of data that AI can process can be overwhelming. AI creates noise and it hallucinates.
Anyone who saw the journalist Sam Coates argument with ChatGPT will know what this looks like.
And so, despite the sophistication of AI, it can’t operate in a vacuum. Human intervention remains essential to:
- interpret results
- validate recommendations
- and ensure that decisions align with organisational values and strategic goals.
AI can sometimes get it wrong. It could be due to biases in the data it’s reviewing, lack of context, or unforeseen variables. Without critical human thinking, these errors can have significant consequences.
This underlines the importance of demonstrating the power of the partnership that should exist between humans and AI.
Each plays to their own strength.
This is urgent. AI needs us, NOW!
One of the most immediate impacts of AI is the explosion of available data. AI systems can ingest, process, and analyse information from countless sources, generating insights at a pace that far exceeds human capacity.
Going back to IBM, their Defence Simulation Analytical Service, or DSaS for short – don’t you just love an acronym! – powered by IBM’s WatsonX, now allows military commanders to access vast amounts of insight, intelligence, and real-time data to simulate scenarios and critical information in seconds. It’s robot wars time!
And while this capability could be transformative, it also produces a massive amount of information overload. So now military commanders risk being faced with more data than ever before and becoming paralysed by the plethora of choice. History has shown us how we can all too easily be misled by irrelevant or misleading insights.
Sharpen up your data literacy and critical thinking!
The ability to filter, prioritise, and contextualise information becomes the core competency in the AI age.
IBM is quick emphasise the importance of developing data literacy and critical thinking skills to navigate this new landscape. As our client Stephen Kelly said in one of his recent posts:
AI-driven work practices will automate document management, scheduling and diaries, complete compliance checks and risk assessments which will reduce administrative requirements for employees. All this will allow employees to focus on the quality of their work which is, according to thinkers like Daniel Pink, a key motivator for all workers.
In our own businesses, we need to tread a fine line. Experiment with AI, stay ahead of the competition and learn how to interpret and deploy it.
Employees must be able to assess the quality and relevance of AI-generated insights, question assumptions, and recognise when additional context or human judgment is needed.
This isn’t just about technical proficiency, but about cultivating a mindset that values curiosity, dare I say it a touch of cynicism, and a dollop of calmness.
The journalist and undercover economist/data warrior Tim Harford sums this up beautifully when he says:
When we’re given a piece of information, we just assume that’s all the information we have. But context is important. Context is “okay, you’ve given me this information, but what other information is out there?” Context could also be, well, what do you mean when you say x?
Humans do empathy far better than AI
As psychologist Daniel Kahnemann points out in his book Noise, there are three factors that compromise human judgement:
- Occasion Noise, or Mood – we're affected by the environment that surrounds us. That can be anything from the temperature (judges hand out different sentences on hot days as opposed to cold days) or who we surround ourselves with
- Level noise, or Personality and Severity – it's where our judgements are inconsistent. We have different attitudes towards decisions and the severity of those decisions (insurance brokers can reach decisions on the same premiums that can differ by 50%)
- Pattern noise, or our Attitude and Values – we all have a different set of values that have been instilled in us over the years. That diversity is essential to making balanced decisions.
But, these perceived weaknesses in the context of AI can also harbour our strengths. Where AI excels at processing structured data and identifying patterns, it often lacks the nuanced understanding of human experience, i.e. empathy.
For example, in HR, AI can automate routine tasks such as candidate screening and onboarding. However, it’s human judgement that is required to evaluate cultural fit, assess soft skills, and make hiring decisions based on whether a team member brings the diverse experience required.
Another example is IBM Watson Health which helps clinicians analyse patient data and treatment options at speed. But it’s the doctor who ultimately makes the diagnosis and treatment plan. They can draw on their medical expertise, their understanding of the patient (their attitude towards treatment) and any ethical considerations.
Learn how to talk to AI. It’s all about language.
The author and poet Kate Clanchy has written a beautiful book on what she has learned teaching English in schools. In it, she explores the lost art of creative writing – the ability to master our words and express ourselves.
Through her work Clanchy explains how we teach the technicality of language (commas, apostrophes etc.) and some of the stories (quite why my boy’s school felt I’d be stirred by Jane Austen is beyond me), but we fail to teach – or nurture – the use of language to express ourselves. The creative writing has got lost.
She examines the unique power of our own human storytelling – that ability to channel deep emotion, cultural insight, and personal experience into words that truly resonate. That process itself is a vital form of thinking and discovery. It helps us foster the critical skills and authenticity that no algorithm can replicate.
It’s also the new programming language. The native tongue and how we use it, matters.
I’ll show my age now. I remember being placed in a classroom with a BBC and then Commodore computer and being handed a floppy disc. Google it.
I then remember being told to write some code to create an address book for all my contacts. The flashing green cursor on the screen invited me to start writing in BASIC. Quite why I needed this laborious address book when all my numbers (and there weren’t many) were scribbled in a notebook, was unclear to me. Either way I did it.
But now, the language and clear prompts we use to direct AI in the right way is critical. Helping our employees understand that matters.
That may mean that a subject like English (insert language as appropriate) is something that our Learning & Development departments need to explore again.
Five actions we should all consider
- Verify everything: Double-check all AI outputs and treating them as suggestions, not final answers
- Explore more: Examine why an AI gave a certain result and explore what assumptions it made, the balance of its thinking, and see what else is out there
- Be critical: Trust in AI’s athletic capabilities but be ready to challenge its results
- Think long term: Consider the implications of decisions, including fairness, privacy, societal impact, and long-term benefit to all those affected by what AI suggests
- Prompt clearly: AI is only as good as the instructions you give it or the data you supply it. Flawed data in equals flawed results out and as much of the data AI trawls can be from unverified sources be sure to give AI what it needs to give you accurate results.
The future relies on Humans empowering AI
Much has been made of how AI will empower us. I think it’s the other way round. We have a hugely powerful tool at our disposal. The operative word there is ‘Our.
Empathy, creativity, and problem-solving skills – this is the age we are moving into. These were once the preserve of the marketing department. Not anymore. Every role will need this skill, and every culture will need to encourage that.
