When we step back and consider what we know of human history, and then even further to planetary history, the emergence of artificial intelligence (AI) at the same time that Earth has crossed over seven of the nine planetary boundaries is without doubt one of the most defining moments of our existence.
This current decade (2020-2030) has already been dubbed as the “decisive decade”. 2030 is not only the target year to achieve the United Nations Sustainability Goals (SDGs) “for peace and prosperity for people and the planet, now and into the future”, it is also the target year to achieve our global goal of reducing greenhouse gas emissions (GHGs) by 45% from 2010 levels (to ensure we met net zero around 2050). It is this goal that will avoid the most catastrophic climate change impacts that will last for centuries to millennia.
It’s now almost the end of 2026, which means we have just over three years to achieve that goal – only now we have Artificial Intelligence (AI).
AI was not in the picture when these goals were set. Now it changes everything.
Whether you like it or not AI is here and it is rapidly changing our world. Climate change is also rapidly changing our world.
As the power and temptation of using AI become harder for business leaders to ignore, the responsibility on their shoulders grows. How we choose to embrace AI over the next couple of years will shape the future of our planet, and the lives of those living on it now and in the generations to come.
There is no denying the power of AI, especially when you combine it with robotics – machines that some are calling “humanoids”. Elon Musk claims these “humanoids” “can be a teacher, babysit your kids, walk your dog…just be your friend…just whatever you can think of, it will do. And yea, it’s gonna be awesome.” – Elon Musk (see from 7:07).
We find that terrifying, and we think it demonstrates AI’s biggest risk – the people behind it.
AI technology has largely been developed by people whose priorities appear to be maximising their profit and colonising other planets, rather than repairing our own planet or eradicating poverty.
Look no further than Elon Musk, Mark Zuckerberg, Sam Altman, and Jeff Bezos. These are people who want to inhabit Mars, live in a metaverse, think the energy used to train AI is comparable to the energy used to “train” a human, or excessively surveil their workers.
Of course, it’s not only the tech founders themselves driving the rapid expansion and development of AI but also those who are investing in their businesses. There is an unprecedented amount of money being invested into AI – forecast to skyrocket to USD$2.5 trillion in 2026. For a comparison, it has been calculated that it would cost roughly USD$318 Billion per year to reduce the world’s poverty rate to 1%.
Those investors expect a return on their investment. That expectation puts real pressure on company leaders to show progress quickly, and considerations about environmental impact, labour practices, and social consequences are already being superseded by the drive for growth and profit. Just take the push to build AI data centres – a Gallup poll found that 7 in 10 Americans oppose local construction of AI data centres and $64 billion of data centre projects have been blocked or delayed amid local opposition. The industry appears to view this opposition as a financial risk, rather than a clear signal that they need to rethink the pace of growth and the scale of expansion. We can’t help but think of the well-known Facebook motto (a major developer of AI) – “move fast and break things”. A recipe for success…?
Our fundamental message about AI is: we can’t let AI just happen to us.
We cannot simply sit by and assume the decisions being made about AI are good for business or good for humanity. Those decisions currently sit with a small group of people who we have little reason to trust will steer it towards an outcome that will benefit anyone but themselves and their investors.
What does this mean for business leaders?
It comes down to how you embrace AI and how you talk about it. Our advice is to understand it and be open about it.
Understand AIs capabilities – use it, test it, experience what it can and can’t do – understand who is behind the tools you’re using and what their intentions are, and understand the impacts it’s having, both positive and negative. Talk to a wide and varied range of your stakeholders about it too – what are their concerns, hopes, current use cases, and understandings of AI? Then take that understanding and feedback to develop a policy that clearly sets out how AI is used, and not used, in your business and how the environmental, social, and governance considerations are weighed. (See below for a list of considerations). Then be open and honest about your use of AI with your stakeholders.
Don’t let AI become some kind of “dirty little secret” within your organisation. Front it and own it.
Finally, stay engaged beyond your own business – follow policy conversations and advocate for regulations that hold AI to the same social and environmental standards we expect of any other business. Be and informed and constructive contributor to the conversations that will shape the future of our economy, our society, and our planet.
For more detailed advice, we are providing highlights from a briefing paper we provided a client to help them consider how they would embrace AI. It is by no means an exhaustive list, but we hope it helps more businesses engage with AI thoughtfully and openly in the ways that we are advocating for.
Environmental Considerations
Water: Data centres use water directly for cooling, and for some, indirectly through the electricity generation that powers them. Many data centres are known to have been built in water stressed areas and compete with local people, ecosystems, and businesses for water. Large data centres can consume up to 5 million gallons per day, equivalent to the water use of a town populated by 10,000 to 50,000 people.
Energy: Data centres use vast amounts of energy. This is energy that needs to be newly generated or taken from existing electricity grids, which in turn places increased demand on the local infrastructure and energy market. From 2024 to 2030, data centre electricity consumption grows by around 15% per year, more than four times faster than the growth of total electricity consumption from all other sectors.
GHG Emissions: AI training and use draws on electricity grids that are largely still reliant on fossil fuels, resulting in increasing data-centre emissions. Constructing data centres and manufacturing hardware carries a substantial carbon footprint also – Microsoft, Amazon and Google’s collective carbon emissions have increased by nearly a fifth in the past year, driven largely by data centre construction.
Resource Extraction & Land Use: Data centres require huge volumes of precious metals (AKA rare earth metals), to construct, which are often mined in environmentally destructive ways. Building data centres requires large tracts of land for server halls, cooling infrastructure, and future expansion, often converting farmland, forests, or wetlands in the process. By 2030, the land footprint of global data centres to exceed 14,500 square kilometres — roughly twice the size of the Jakarta metropolitan area (home to more than 32 million people).
Social Considerations
Jobs: Will your use of AI create jobs or take jobs? This question should be considered not just for full or part-time employees, but also for the contractors and temporary workers (e.g. models or talent for advertisements, videographers, editors, etc). Consideration should also be given to career pathways. As many of the tasks AI is commonly used for are entry-level type tasks, what impact might that have on attracting new talent and providing pathways into the business for younger people?
Professional Development: What impacts could the consistent use of AI have on the professional (and personal) development of your people? Growing reliance on AI for thinking and decision-making tasks may be reducing critical thinking, memory, retention, and independent problem-solving – an MIT study had participants write essays using ChatGPT, Google search, or neither, while tracking brain activity via EEG. ChatGPT users showed the lowest brain engagement, underperforming the other groups on neural, linguistic, and behavioral measures.
Productivity and Wellbeing: Assuming AI makes an employee significantly more productive, who benefits from that gain? Does the higher productivity simply become the expected norm with hours unchanged? Or could that employee choose to work fewer hours (with or without an impact on their pay), and spend that time with their family and friends, in their communities, restoring nature, looking after their wellbeing, or volunteering?
DEI: How could using AI impact biases in your hiring process? AI can replicate and scale discrimination in hiring, lending, and policing decisions –– the most comprehensive independent study of AI-powered hiring algorithms ever conducted has found stark racial disparities embedded in the tools used to screen millions of job applicants, with more than one in four applications submitted by Black job seekers directed to positions where the algorithm produces outcomes that trigger federal discrimination scrutiny.
Governance Considerations
Culture: Consider how AI use might affect your internal culture, both positively and negatively. Are your people open about how they use AI, or is it becoming a “dirty secret”? And are your people talking to AI rather than each other? Creating a culture of responsible use of AI and having a broader positive team environment requires openness, trust, and collaboration.
AI Entrapment: AI tools are currently very cheap and deliver significant value, which makes it tempting to embed throughout a business’s operations. However, the companies and investors behind these AI tools will eventually expect them to return a profit. If a business has become too heavily reliant on AI and then prices are dramatically increased, then it may have few alternatives, especially if skilled staff and in-house knowledge has been lost along the way.
Brand: Being transparent and authentic about how you use AI and its impacts will build trust and credibility with customers, and other stakeholders. In practice, this could include disclosing when any content has been created or revised using AI, or when AI has played a role in developing a product or service.
Competitive Advantage: Given the mass adoption of AI by businesses across the world, simply using it is unlikely to set you apart. How can your approach to AI use become a point of difference?
Human Rights & Modern Slavery: How will your use of AI impact people in its supply chain? Training AI models is often outsourced to low-wage labour in developing countries under poor conditions and pay. A May 2025 Equidem report on 116 data-labeling and content moderation workers in Colombia, Kenya, and the Philippines (working for firms serving Meta, ByteDance,Remotasks, and OpenAI) found widespread labour rights violations: unsafe exposure to traumatic content causing severe mental health harms, workdays of up to 20 hours, unstable pay and job security, retaliation against workers who spoke up, and suppression of unionizing efforts via NDAs and intimidation. Reporting on this could become a requirement of anti-modern slavery legislation.
More than ever right now the planet needs help to repair the damage we have done to it. We need to get plastics out of the ocean, GHGs out of the atmosphere, water back in our wetlands, and trees back in the ground. AI could help with this practically, and by allowing people en masse more time to do the physical work required.
There is no doubt of the power of AI and the good it does and could do. But there is equally no doubt of the harm it does and could do. AI is a reflection of how it has been trained and, for now, those choices are still in the hands of people with questionable motives and character with little to no regulation to ensure the worst-case scenarios do not become reality. Likewise, there is little to no regulation on how AI is used. Let’s not leave it that way. AI shouldn’t just happen to us; we all have a role in shaping what it becomes.
You can read our AI policy here and our decision to shift from Chat GPT to Claude here.
——–
AI Disclaimer: AI (Claude) was used for some of the research cited in this article. It was not used to construct or review any part of this article.
Written by Nick Morrison, Founding Director at Go Well Consulting.
Image by Cotton Bro Studio