AI is a fantastic tool for so many things. Whether checking you’ve covered every angle in a sensitive message, brainstorming ideas to spark your creativity or optimising content for search engines, platforms like ChatGPT and Microsoft Copilot are incredibly useful.

With every day that passes, their capabilities grow exponentially – and now they’re supercharging top professionals’ creative processes.

Fortune business editor Nick Lichtenberg made the news in March for producing more stories in six months than his colleagues write in a year. How? By using AI in creative ways, including:

  • Analysing massive technical documents to find newsworthy lines
  • Gathering sources that backed up factual information in his writing
  • Developing structures for his articles

Lichtenberg told Reuters his boss had summed up the scale of his AI use by saying most of his stories were ‘at least 50% me’.

So not only is AI upping productivity, but it’s now sophisticated enough to be the fundamental tool behind a leading journalist’s content. Elsewhere, business owners are already realising its potential to save on overheads, with a number of major companies recently citing AI efficiencies as a contributing reason for staff layoffs.

But before committing to AI as a fundamental part of creative processes, we all have a responsibility to consider some important practical, ethical and societal questions.

Is AI sustainable?

It’s easy to ask AI a question without considering how it provides an answer. Every AI interaction uses energy through ‘data centres’ – giant computer systems that consume massive amounts of electricity and water. According to Coventry University, as of June 2025, two per cent of the world’s power was already being consumed by AI – and it could be using 1.7 trillion gallons of water each year by 2027. That necessitates high rates of fossil fuel and natural resource consumption.

And even if the world was able to power AI entirely with renewable energy, its hardware, data and power storage infrastructure require the mining of precious minerals.

The London School of Economics and Political Science writes of the environmental consequences, including deforestation, soil erosion and water and soil contamination. If your organisation has green environmental policies and a commitment to sustainability, it’s worth asking yourself whether they’re reconcilable with upping your AI use.

Is AI going to get more expensive?

At the moment, AI solutions are usually cheap and, in some case, free. That’s because AI firms have effectively been providing subsidised products, spending huge amounts of capital to gain market share without recouping their investments. Sooner or later, they’ll expect a return, which may come when AI is so integral to business that going without is no longer an option.

Costs could also ramp up as AI systems become more sophisticated – complex tasks require additional energy, as well as specialist human maintenance and operation.

Will AI stunt the development of the next generation?

Educationalists already accept that students are going to use AI to complete their work. The Higher Education Policy Institute (HEPI) last year found 88% were using tools like ChatGPT for assessments. While the most common uses were to explain concepts, summarise articles and brainstorm ideas, a quarter of students surveyed directly used edited AI copy.

Studies into how AI can support school-age learning are also ongoing. At all ages, though, academics have expressed concern that unless AI is developed to complement, not replace, interactive learning, future generations risk problems with memory retention, creativity, critical thinking and problem-solving skills.

Fact-checking AI is a specialist skill in itself

AI is good, but it’s not that good yet. It’s still prone to spewing out misinformation in the form of ‘hallucinations’ (made-up stats based on assumptions) or ‘bias’ (failing to account for unfairness in data). Yes – Nick Lichtenberg has leant heavily on AI, but journalistic skills curated over decades at major publications have still been crucial.

A major turning point happened when AI summaries started citing their sources, allowing Lichtenberg to more easily establish their accuracy. And like any other journalism, his work was thoroughly fact-checked and edited by skilled newsdeskers before publication.

Without forensic checks and balances, AI users run the risk of making errors that could affect reputation, business decisions and the quality of advice or services offered to clients.

Conclusion

In just a few years, AI has graduated from being a tool of convenience into something much more meaningful. Its capacity to exponentially boost output at a relatively low cost is already fundamentally influencing how businesses operate.

However, it’s important to remember that developers will eventually need a return on their enormous investment – and that’s when the cost of AI may increase. There are wider anthropological – and even existential – questions around AI that are likely to necessitate major research, regulation and legislation before long.

With so many unknowns, approaching AI with a healthy dose of scepticism is important – a collaboration between AI and the human brain is still essential, for now. Failing to capitalise on the benefits AI offers runs the risk of being left behind. However, those benefits need to be considered against your organisational values. And there are major pitfalls attached to over-reliance on AI in the creative process, including the risk of publishing misinformation – it’s prudent to proceed with caution, scrutiny, checks and balances to avoid them.

Share