Quarterly Outlook
Upending the global order at blinding speed
John J. Hardy
Global Head of Macro Strategy
Saxo Group
A few years ago, artificial intelligence sounded futuristic. Today, it has moved into financial markets, enterprise software, logistics, consumer electronics, healthcare, and virtually every sector of the economy. Companies are not only integrating AI into existing operations but also rebuilding products, business models, and investment strategies around it.
Investors have taken notice. AI-related stocks have soared in recent years. Tech giants have shifted capital expenditures into AI infrastructure and model development. Startups have raised billions to compete with established players in the market while governments and regulators scramble to understand the implications of this rapid shift.
Despite the excitement, the long-term viability of investing in artificial intelligence remains a complex question. Some companies may dominate, while others struggle to justify their valuations. Ethical and data privacy concerns, as well as geopolitical tensions, also add layers of risk that investors cannot ignore.
AI has transitioned from a niche technological interest to a central focus in both public and private markets, driving significant capital shifts over the past decade. From foundational model developers to cloud infrastructure providers, nearly every layer of the digital economy is being revalued through the lens of artificial intelligence investing.
Private investment in artificial intelligence has seen a substantial rise in recent years. Funding for AI and cloud-focused companies across the United States, Europe, and Israel has grown significantly, with generative AI emerging as a major area of interest—accounting for a notable share of total investment. The United States has been at the forefront of this trend, attracting the majority of capital directed towards generative AI ventures. Prominent companies such as OpenAI, xAI, and Anthropic have played a key role in driving this momentum.
AI investment trends have also significantly impacted public markets. Leading AI stocks like Nvidia, Microsoft, and Amazon have experienced substantial gains, driven by increasing demand for AI chips, cloud computing services, and enterprise AI adoption. For instance, Nvidia's market valuation surpassed USD 1 trillion in 2023, reflecting strong investor confidence in AI infrastructure and chip technologies. Venture capital continues to flow into startups developing AI models, infrastructure, and applications, with many positioning themselves as challengers to current market leaders.
The scale of the opportunity is a major driver. Analysts project that the global AI market will expand to over $826 billion by 2030, reflecting robust adoption across various industries. Concurrently, the soaring demand for data centre capacity, computational power for AI model training, and skilled AI engineering talent will continue to attract institutional investors, hedge funds, and retail investors to this rapidly evolving theme.
As a result, investing in artificial intelligence is increasingly viewed not merely as a technology-sector play but as a foundational theme across industries. Thematic portfolios, AI-focused ETFs, and index-tracking products now offer investors multiple entry points into a fast-moving and competitive landscape.
AI is no longer a niche subsector of tech. It is embedded across infrastructure, hardware, software, and services, shaping the investment landscape far beyond Silicon Valley. For investors evaluating how to gain exposure, there are several pathways.
Some of the highest-profile opportunities lie in listed companies leading core AI innovation or benefiting from its widespread integration.
These include AI hardware manufacturers, and firms that offer cloud-based services, custom chips, and language model development. Other major players in the space include tech companies that embed generative AI into search, productivity, and user-facing products, as well as organisations that leverage AI to optimise workflows, security, and content creation.
Investors are also paying attention to AI chipmakers and networking companies, which have been experiencing rapid growth due to increasing infrastructure requirements of AI model deployment.
Some of the biggest names in the AI space are:
For those seeking broader exposure, AI ETFs and thematic funds offer diversified entry points. These portfolios reduce company-specific risk and allow investors to capture general AI adoption trends without betting on a single winner.
Other funds also track indices focused on AI-adjacent technologies, such as semiconductors, automation, or cloud computing, offering a lower-volatility path to AI exposure.
While the public markets reflect the dominance of Big Tech players, private investment continues to fuel AI startups. Companies building specialised LLMs, vertical AI tools (e.g., legaltech, medical diagnostics), or edge AI solutions have raised billions in venture capital.
These challengers often develop open-source or fine-tuned models that compete directly with giants like Google and Meta.
For most investors, exposure to these firms is limited to indirect routes (e.g., VC-backed IPOs or through holdings in larger companies that acquire or partner with these startups). However, their growth trajectory often shapes the valuations and strategies of public AI stocks.
Investing in artificial intelligence requires looking across the AI stack, from data centres and chipmakers to software platforms and enterprise users. No single segment captures the full set of opportunities. Diversifying across infrastructure, model development, and applied AI could help you balance short-term volatility with long-term growth potential.
The rise of AI has triggered a wave of investor interest, but identifying sustainable winners requires more than just recognising brand names or recent share price gains.
Here's what you should pay attention to:
AI products often benefit from strong economies of scale. Language models, once trained, can be repurposed across platforms and industries. However, training large models requires immense capital expenditure and computing power. Investors should evaluate whether a company can scale cost-effectively, either through proprietary infrastructure or partnerships that reduce marginal deployment costs.
Access to unique datasets is a critical differentiator because AI systems improve when trained on relevant, high-quality data. Companies with exclusive datasets, such as healthcare firms with anonymised patient records or financial institutions with real-time transaction flows, have a structural advantage. This data edge makes their AI offerings harder to replicate.
Not all AI solutions are equal. Some firms embed AI into existing tools to improve user experience, while others build AI-native platforms that offer entirely new services. A clear product roadmap, strong user engagement, and defensible IP (such as model architecture or deployment infrastructure) often signal that a company can maintain a lead even in a competitive landscape.
Many AI startups attract attention with impressive demos and early adoption, but long-term viability depends on sustained revenue growth. Sustainable AI investing means scrutinising how a company captures value (via subscriptions, usage-based pricing, enterprise integrations, or licensing). Clear pricing tiers, customer stickiness, and high-margin recurring revenue are strong indicators of monetisation maturity.
With AI markets evolving rapidly, execution matters. Does the company ship updates regularly? Has it translated R&D into customer adoption? Are partnerships expanding its market reach? Earnings reports, user growth metrics, and product releases offer tangible signals of execution strength.
AI companies' role in the ecosystem (model builder, infrastructure provider, platform integrator, or application developer) affects their growth ceiling and margin profile. Investors should identify whether a firm is leading in its segment or facing pressure from larger, more integrated players.
Investing in artificial intelligence isn't just about growth potential anymore. Ethical risks are now front and centre for investors who care about long-term value and impact. Here are the main issues to watch closely:
AI systems often learn from real-world data, but that data isn't always fair. If the training data includes past inequalities or imbalances, the AI can repeat them in ways that hurt people. This includes hiring tools that prefer one demographic over another or healthcare models that don't work well for specific populations. Investors should ask whether companies have built-in checks, such as independent audits or teams focused on responsible AI.
AI systems frequently rely on large-scale data collection. This raises significant concerns around data misuse, surveillance, and consent, particularly in sectors like retail, advertising, and biometric recognition. Poor data governance practices can lead to legal liability, reputational damage, and regulatory fines. Investors should consider whether firms are compliant with GDPR, CCPA, or other major data protection frameworks.
Many AI systems learn by analysing content scraped from the internet, such as books, code, music, or news articles. But that raises big questions: was the data used legally? Could the AI generate something that violates someone else's rights? Companies being sued over this are already making headlines. Safer bets are firms that are open about what data they use and have clear licensing policies.
Some AI systems work like black boxes—nobody really knows how they decide things. That's a problem when they're used in areas like healthcare, lending, or law enforcement. Investors should look for companies that explain how their AI works, allow third-party checks, or share parts of their technology openly.
AI needs serious computing power, which means it uses much energy. Training a large model can emit as much carbon as multiple cars over their lifetime. Investors who care about sustainability should check whether companies are using efficient systems, green energy, or carbon offsets.
Some AI tools are used for surveillance, predictive policing, or even autonomous weapons. These 'dual-use' cases may raise red flags for investors focused on ethical governance. Check whether the company has clear policies on how its AI can and can't be used.
Artificial intelligence is reshaping the job market. Some roles are being streamlined or eliminated, while others are evolving. For investors, this shift presents both a business and a reputational risk.
AI-led automation has impacted jobs that involve routine, repetitive work. Manufacturing, logistics, customer support, and admin roles are increasingly handled by robotics, chatbots, and workflow automation. In finance, tasks like compliance checks and basic reporting are now partly automated, reducing the need for certain back-office functions.
Creative and professional jobs are also feeling the impact. AI tools can write code, generate marketing content, review legal documents, and assist in medical diagnostics. While these tools boost efficiency, they also lower the demand for entry-level roles in design, media, law, and healthcare.
But not all effects are negative. AI is also creating new roles in data management, model oversight, and technical operations. Many companies are reshaping their teams, not just shrinking them.
Governments are watching closely. Some are testing policies like AI risk assessments, automation-linked taxes, and retraining incentives. As public concern grows, companies that automate without supporting workers may face regulatory pressure or reputational fallout.
For investors, the key question is whether companies are handling workforce changes responsibly. Long-term success will depend not just on cost savings but also on how sustainable and socially viable their automation strategy really is.
AI is one of the most transformative themes in global markets, but it also brings risks that investors need to understand before allocating capital. Here are the main ones:
1. Valuation and sentiment riskSome AI-related stocks have seen dramatic price increases, particularly in semiconductors, cloud infrastructure, and automation software. But many of these gains are driven by future expectations rather than current earnings. This creates exposure to sharp corrections if investor sentiment cools down, earnings disappoint, or interest rates rise. That's why betting heavily on the most popular AI stocks can make a portfolio more volatile.
2. Changing rules and regulations
Governments around the world are still figuring out how to manage AI. New rules about privacy, safety, fairness, or how data is used could change how companies operate. If a company doesn't follow the rules or can't adapt quickly, it could face fines, delays, or negative headlines. Since different countries have different rules, global AI companies may face extra challenges.
3. Execution riskNot every company marketing itself as "AI-driven" has viable, scalable solutions. Some firms struggle to move from research and prototypes to real-world applications that generate sustainable revenue. Business models can break down if integration costs run too high or if customers are slow to adopt. Investors should assess not just the technology, but its actual commercial impact and user traction.
4. Relying too much on a few providersMany AI companies depend on the same big players for chips, cloud services, or key technology. If one of those providers raises prices, cuts access, or faces political problems, it can affect many smaller companies at once. This kind of dependence can create hidden risks.
5. Data dependency and model risk
AI systems are only as good as the data they're trained on. Poor data quality, outdated inputs, or biased datasets can lead to flawed outcomes, reputational damage, or regulatory pushback. Additionally, model performance can degrade over time ("model drift") or fail under unexpected market or user behavior. This adds a layer of unpredictability to long-term value creation.
6. Talent scarcity and operational bottlenecks
Advanced AI development requires highly specialised talent, which remains limited and expensive. Companies that can't recruit or retain top engineers and researchers may fall behind in innovation or face execution gaps. This risk is particularly acute for smaller or non-core tech firms trying to compete in a field dominated by tech giants.
Some AI-linked stocks have delivered strong returns, but gains have been concentrated, and valuations of leading names are high. At these levels, selectivity matters.
A sensible approach is to spread exposure across different parts of the AI value chain, such as chips, cloud, enterprise software, and industry-specific applications, without overloading any single theme. ETFs can help, but some are heavily weighted toward just a few companies.
Not every company using AI will generate durable growth. So, it's preferable to focus on business fundamentals: how products are monetised, whether customers are sticking, and how efficiently capital is being deployed.
AI remains investable, but it's no longer cheap or simple.