AI in the Office - How ChatGPT is Reshaping the Work We Do

The Houses of Parliament. London - in the style of Gustav Klimt. Antony Slumbers & Midjourney

AI in the Office - How ChatGPT is Reshaping the Work We Do

A revolution was already underway but now the turbo chargers have kicked in. We need the right real estate for this 'new work'.

For at least five years I’ve been writing that we are asking the wrong fundamental question about the future of work. Mostly the focus has been on where we work, and how we work, whereas what is most important is the work we do. 

It is the changing nature of the tasks involved in our jobs that will be the causal factor in how the office, or workplace, of the future is designed, equipped and operated. Where we work, and how we work, will change because the work we’ll be doing is changing.

And the flywheel behind that change has just been hyper scaled. The arrival of ChatGPT, and other derivatives and flavours of GPTs (Generative Pre-trained Transformers) is an extraordinary technological leap. Bill Gates, on the 21st March, wrote that he believed it was the one of only ‘two demonstrations of technology that struck me as revolutionary’, the first being the graphical user interface, ‘the forerunner of every modern operating system, including Windows’.

So this is a moment in time to note. A pivotal moment.

Why? Because GPTs enable advanced AI-driven natural language understanding and generation, which significantly enhances human-computer interactions. This innovation empowers businesses to automate tasks, optimise decision-making, and streamline customer experiences in ways that were previously impossible. This will drive efficiency and innovation across most if not all industries. At great scale and with great speed.

I used to talk about how any task that was ‘structured, repeatable, predictable’ would be taken over by ‘machines’. And this has increasingly been the case. Just look at the rise of ‘Robotic Process Automation’ over the last few years. Anything entered into a computer via a set order of keystrokes can easily be automated. And much has been.

But what GPTs are set to enable is many orders of magnitude greater than any RPO software can achieve. This technology can easily handle whatever is ‘structured, repeatable, predictable’ but can go way beyond this. 

For example, in customer service it can automate responses to routine inquiries, enhance the efficiency of helpdesks, and reduce resolution times. It can assist marketers in content generation, social media management, and targeted messaging. Within human resources it can aid with streamlining recruitment processes, employee onboarding and training, by automating repetitive tasks and providing personalised learning materials. R&D can be accelerated by improved ideation, trend analysis and knowledge synthesis. Sales in turn can be supported by automating lead generation, nurturing, and follow-ups.

Frankly this is the tip of the iceberg. GPT researchers are finding that this technology has capabilities beyond what they imagined, or envisaged. There seems to be something about the processing of language at massive scale, alongside vast quantities of training data, that is pushing the boundaries of what was thought possible in computing. A report from Microsoft Research says this in its abstract: ‘We demonstrate that, beyond its mastery of language, GPT-4 can solve novel and difficult tasks that span mathematics, coding, vision, medicine, law, psychology and more, without needing any special prompting. Moreover, in all of these tasks, GPT-4’s performance is strikingly close to human-level performance’.

So, a lot of what we did at work we’ll not be doing in the future. We must reclassify many tasks as ‘Old Work’. Work that we will off-load to machines.

Way back in 2017 McKinsey wrote that they believed 49% of the work people were paid to do in the global economy could be automated by adopting currently demonstrated technology. That was then, this is now. With technology far superior to what we had available in 2017.

Which leaves us humans with our own major pivot to make. We urgently need to concentrate far more than we have done on those skills that humans possess, and machines, even GPTs, do not.

I wrote about this as ‘New Work’ - work that required the distinct capabilities of humans. Which broadly speaking are design, imagination, inspiration, creation, empathy, intuition, innovation, abstract & critical thinking, collaboration, social intelligence and judgment.

Let’s think of these in the context of five ‘workflows’ that are likely to represent much of the ‘work we do’ in the future:

1. Critical thinking and problem-solving: Evaluating AI-generated suggestions, making well-informed decisions, and identifying complex issues that require human intervention.

2. Creativity and innovation: Leveraging AI as a tool to develop new ideas, products, and strategies while thinking, as humans, about value propositions and competitive advantage,

3. Emotional intelligence: Understanding and managing emotions, empathising with customers and colleagues, and navigating interpersonal relationships effectively. Something which AI cannot fully replicate.

4. Adaptability and learning agility: Embracing change and quickly acquiring new skills in response to evolving business and technological landscapes.

5. Domain expertise: Possessing deep knowledge of specific industries, regulations, and best practices to contextualise AI-generated content and ensure compliance and accuracy.

These are deep rabbit holes, where AI needs to be co-opted to augment our unique capabilities as humans.

Picasso nailed it decades ago when he said ‘Computers are useless - they can only give you answers’.

Humans are here to ask the right questions. That in itself might be the super skill of the future; the ability to ask the right questions.

So we have ‘Old Work’ and ‘New Work’.

The real estate issue is that much, perhaps even most, of our offices, our places of work, are designed for ‘Old Work’. Which is in the process of leaving the building.

All of this space is, or shortly will be, obsolete.

The future proof office has to be designed around ‘New Work’. It has to be somewhere where human skills are catalysed. Somewhere conducive to critical thinking and problem-solving, creativity and innovation. Where emotional intelligence is prized and adaptability and learning agility is the default setting.

Space that is at the service of these human skills.

#SpaceasaService - where humans thrive in a world of GPTs.