The power and the profit potential of AI are mesmerizing, and the equity markets have certainly benefitted from the dream.
What do we need to consider in 2025? Will the surprises continue? To help train ChatGPT’s response, we analyse the actual revenue that has emerged so far, and the revenues that could emerge in the future.
Revenue potential: Getting realistic
Since the launch of ChatGPT[1], artificial intelligence equities outperformed the broader market by almost 70% (Figure 1). It isn’t an understatement to claim that AI stocks were the most important driver behind the positive global market performance.
Let’s consider what AI revenues actually exist today, and what revenues we will likely see in the future. The sweet spot in 2024 was companies such as Nvidia and Taiwan Semiconductors, that is, companies which train AI models to make predictions or conclusions – and, of course, their suppliers. These range from specific types of semiconductors to data center cooling equipment and even some electricity generators. For these types of companies, tangible revenue from AI already ranges from 10% to 80%[2] of their total, they have already enjoyed tremendous share price performances. Based on the capital spending plans of the AI data processors, we expect this sector will continue to enjoy tremendous revenue growth.
Runner-up companies, such as IT consultants and other data management sectors, are beginning to gain traction in AI revenues. Examples are Accenture consulting and Servicenow software. Their potential clients in several sectors hold large amounts of data which these clients are (as of yet) unable to leverage in an AI framework. For the third category, including the clients of the consultants, it may be too early to know which end users will be able to use AI to improve their margins. Drug discovery, industrial production processes, and financial firms will likely be among the beneficiaries of AI, if not in revenue, then in cost savings. This could be a huge, and diverse, list – think Holologic, SAP, and even Candriam!
Is ChatGPT teaching itself about the risks?
We believe that AI will be a powerful driver of the global economy. It is always difficult to predict the precise time frames of new technology. For example, will the new supply of power keep up with the massive demand for electricity used by AI? The consulting firm McKinsey estimates that power for data centres in the US could be as high as 12% of the total US power demand.[3] At the moment there is a shortage of skilled AI experts, and this scarcity might hamper full AI deployment.
Past performance does not guarantee future results.
We believe the greatest risk to growth of AI is abuse, even if unintentional. Biases can result. For example, poorly ‘trained’ lending algorithms that offer worse loan terms or deny credit to minority applicants. Misinformation can be multiplied using AI-powered tools, such as ‘deepfakes’. During the Covid-19 pandemic, AI-driven algorithms on social media amplified vaccine misinformation. Data privacy breaches are also an obvious challenge.
As individuals, we are all stakeholders in these AI risks, even if we choose not to be investors.
Facial recognition technology (FRT), one of the segments within AI, allows us to board planes more quickly and offers convenient security for our mobile phones. Conversely, inappropriately trained FRT has resulted in false arrests of innocent people, particularly minorities. Some authoritarian countries use cameras and facial recognition technology on their citizens to collect personal information without consent. With one billion surveillance cameras in use globally by the end of 2021, we consider FRT to be the highest-risk use of AI.
Our future history: Placing the guardrails
It is up to stakeholders, especially investors, to help establish the guardrails. In 2021, Candriam led an FRT investor statement signed by 55 investors representing over $5 trillion in assets under management. Industry leaders took note, and some took action. As investors increasingly consider our responsibilities in financing AI, asset managers representing $8.5 trillion signed the 2024 Investor statement on Ethical AI, co-led by Candriam.
Regulators in different countries are pulling in opposite directions. In the EU, the Artificial Intelligence Act phases in a risk-based framework, in an effort to balance the benefits of this technology with public safety. In the US, proposed members of Donald Trump’s incoming administration, such as Elon Musk, have advocated for reducing regulatory barriers in tech. In the absence of strong regulatory frameworks around the world, the responsibility for ensuring ethical AI practices increasingly falls to businesses and investors
Oh, Baby it’s a Wild World!
The World Benchmarking Alliance says, “Both the risks and opportunities of AI have materialised with exceptional speed in the last two years.”[4]
AI is accelerating scientific research in healthcare, improving our infrastructure and communications, and helping us develop the urgently-needed new technologies to tackle climate change.
Investors can help limit both our own investment risks, and societal risks, by being selective. We can emphasize investments in companies which contribute positively to a safer AI deployment, such cyber security companies or those which enhance privacy.
It is our Conviction that companies which embrace sustainability-related opportunities and challenges in combination with financial opportunities and challenges are the most likely to generate value. We look forward to the future of AI.
[1] Early demo released November, 2022.
[2] Candriam estimates.
[3] AI’s power binge
[4] The World Benchmarking Alliance is the most widely-recognized standard setter for AI, both by companies and investors, and “hosts’ the 2024 Investor statement on Ethical AI | World Benchmarking Alliance