The ethics of AI 

Artificial Intelligence (AI) is clearly having profound implications on the broader economy in the coming years, but will it be a net positive for society?  

While AI has the potential to revolutionise many areas of society for the better, it also has the potential to be misused and have negative side effects.   

Investment and growth in AI are huge and its impact will continue to integrate across broad areas of the economy, so understanding the environmental, ethical and economic consequences will be essential.  

See also: AI: Reality, risk and returns in the next tech revolution

The positive potential of AI  

AI can help to increase productivity and efficiency in a variety of industries, from manufacturing to finance.   

By automating repetitive and dangerous tasks, AI can free up time for workers to focus on more creative and high-level work. It also has the potential to greatly enhance scientific research and development, as it can analyse complex data sets and help researchers make new discoveries.  

For example, enhanced algorithms can analyse large amounts of medical data to identify patterns and provide more accurate diagnoses, improving healthcare outcomes.   

AI can also provide personalised learning experiences and assist with education, helping students to learn at their own pace and in a way that works best for them.   

Overall, the potential benefits of AI are vast, and it is an exciting technology that has the potential to greatly improve many aspects of our lives.  

The risks and unintended consequences   

On the other hand, there are risks. At the catastrophic end of the scale, some experts have warned that superintelligent AI could pose an existential risk to humanity if not properly managed. Setting this aside, there are still plenty of more immediate and significant risks that need managing.  

AI and automation have long been discussed in the context of replacing manual jobs, which are often low in satisfaction and high in physical labour. However, the risk of automation has now extended to white-collar jobs; in fact, this is where AI has the potential greatest impact. We have already seen how software development has, in a very short time, shifted from having a large labour shortfall to being viewed as one of the most AI-threatened professions.   

Many white-collar jobs may become more commoditised, leading to economic and social challenges.  

Energy consumption is also a major environmental concern. Data centres are currently estimated to account for about 1.5% of global electricity consumption, and that’s projected to double over the next five years.   

Another prominent risk is cyber security. Bad actors have sought to undermine institutions and individuals’ online presence for years, and AI is now increasingly being used for cyber attacks and fraud.  

We also observe that there is a paradox that while AI is meant to be about democratising expertise and intelligence; it also seems to be leading to a further concentration of power and control in the largest corporations in the world. This too can lead to negative societal impacts.  

See also: Is investing in AI still too much of a gamble?

Nuanced real-world impacts: not all risks play out as expected  

Many of these potential risks are not black-and-white issues. For example, there had been concerns among radiologists that AI’s ability to analyse X-rays would lead to significant job losses, but what we’re seeing so far is a productivity improvement, with radiologists’ time freed up to do higher value work.   

Likewise, although the energy consumption of AI is huge, much of the additional capacity is likely to be met by renewables. A lot of the data centre rollout is relying on renewables for power supply, not for political reasons, but because it’s the cheapest, fastest way of deploying – making it less susceptible to Trump’s removal of subsidies.   

Solar energy in particular is likely to benefit from huge demand from developers of data centres. Owners and operators of these data centres want three main things from their electricity supplies: power delivered soon, available all the time, and with low carbon impact. No single energy source ticks all these boxes without caveat, but solar combined with battery storage is the clearest winner on these metrics.  

The need for stronger governance and ethical oversight  

There are clearly many ethical considerations around AI’s growth that need addressing. Startlingly, research from Thomson Reuters Foundation found that only one fifth of companies had any sort of governance in place for the ethics of how they use AI, so this is one area that needs significant development.  

We’ve begun to engage with companies involved in AI to explore the governance they’re putting in place.   

Balancing opportunity and risk in the AI era  

The potential of AI is both exciting and daunting. It represents one of the most significant technological breakthroughs since the founding of the internet but – as with the internet – AI has the potential to do both great and terrible things.  

Balancing the benefits and risks of AI will be crucial in shaping a future where technology serves humanity’s best interests.