The artificial intelligence (AI) sector has surged into an era of unprecedented acceleration, marked by meteoric growth in usage and investment while grappling with staggering costs and intense international competition.
What initially emerged as scattered developments has morphed into a sophisticated, high-stakes ecosystem. Traditional monetisation strategies are being reinvented in real time, often at the expense of massive cash burn.
Layered on top of this is a growing geopolitical rivalry, particularly between the US and China, which is shaping the global AI landscape.
The speed of AI adoption and user engagement is now eclipsing the early internet boom, with machines advancing faster than human capabilities. This exponential growth is vividly reflected in skyrocketing capital expenditure (CapEx) trends that show no sign of plateauing.
Also Read: AI power shift: How geopolitics and innovation are rewriting global rules
Startups are fuelling the pace with aggressive product rollouts, capital raises, and innovation, while tech behemoths are reallocating their cash flows to artificial intelligence to maintain market dominance and ward off new entrants.
Capital and infrastructure surge powers AI boom
Massive user uptake: OpenAI’s ChatGPT exemplifies AI’s mainstream adoption, growing its weekly active users by over 200 per cent year-on-year to hit 350 million by December 2024. Since its launch, ChatGPT has seen an eightfold increase to 800 million users in just seventeen months.
Soaring CapEx among Big Tech: The so-called “Big Six” US tech firms (Apple, NVIDIA, Microsoft, Alphabet, Amazon (AWS), and Meta Platforms) pushed their collective CapEx up by more than 63 per cent year-on-year, totalling US$212 billion in 2024. This increase is largely driven by demand for AI model training and deployment infrastructure.
Data centres as AI factories: Data centre spending has surged to US$455 billion globally in 2024, with projections pointing to continued acceleration. These facilities are becoming “AI factories” as hyperscalers and AI-first companies invest billions to scale computational capacity.
Ecosystem expansion: The NVIDIA AI ecosystem has seen exponential growth: 6 million developers (up 2.4 times), 27,000 startups (up 3.9 times), and 4,000 GPU-enabled applications (up 2.4 times) by 2025.
Monetisation: A multi-pronged strategy
Consumer subscriptions lead the way: Flagship AI models such as ChatGPT, xAI’s Grok, Google’s Gemini, Anthropic’s Claude, and Perplexity are monetising primarily through subscription models for individual users.
API and generative search monetisation: Anthropic’s revenue soared more than 20 times to US$2 billion annually in eighteen months. Meanwhile, xAI’s generative search offerings are set to achieve significant revenue growth by 2025.
Enterprise-driven growth: Companies are embracing AI for top-line growth. Glean, which provides enterprise search and AI agents, grew its annual recurring revenue (ARR) more than 10 times to US$100 million within two years.
Integrated AI platforms: Incumbents are embedding AI across entire product suites. Microsoft’s AI division surpassed a US$13 billion annual revenue run rate in 2024, marking a 175 per cent increase. Its Copilot tool is being widely deployed across services. Meanwhile, TikTok has rolled out Symphony, a suite of AI-powered advertising tools.
Specialised AI software flourishes
Industry-specific AI applications are gaining traction:
Software engineering: Anysphere Cursor AI’s ARR surged from US$1 million to US$300 million in just over two years.
Legal services: Harvey hit US$75 million in ARR by April 2025.
Also Read: Southeast Asia steps up: Complexity, opportunity, and the post-China trade shift
Customer support: Decagon expanded its ARR tenfold in a single year, from US$1 million to US$10 million, reshaping customer service into AI management roles.
The cost of innovation: Burn rates and bottlenecks
Escalating expenses: Training frontier large language models (LLMs) is among the most capital-intensive ventures in history, with compute expenses running into the billions. OpenAI’s 2023 compute spend alone was estimated at US$5 billion.
Monetisation vs profitability: As inference costs per token decline, AI becomes more accessible. However, this brings uncertainty to monetisation models and casts doubt on long-term profitability for model providers.
High-burn dynamics: The prevailing equation in the AI world is “High Revenue + High Burn + High Valuation + High Investment”. Collectively, leading private AI model firms (OpenAI, Anthropic, Perplexity, and xAI) have raised approximately US$95 billion to date, against an estimated combined revenue of just US$11 billion annually as of May 2025.
Energy as a limiting factor: AI’s colossal energy demands are causing data centres to rival traditional heavy industries in consumption. The sector’s growth is increasingly constrained by energy availability, not data or algorithms, with grid strain becoming a critical bottleneck.
The geopolitical chessboard of AI
Open-source disruption: The proliferation of open-source AI is undermining proprietary monetisation strategies by enabling “frontier-level” innovation without billion-dollar budgets. This democratisation could commoditise certain capabilities, posing a threat to incumbents.
China’s rapid AI ascent: China is intensifying its AI efforts in strategically vital areas like battlefield logistics and autonomous systems. By Q2 2025, it had released three open-source LLMs and is closing the performance gap with US models at a remarkable pace, a stark contrast to its late adoption during the internet era.
Strategic tug-of-war: The US and China now view AI as both an economic engine and a geopolitical lever. American policymakers are tightening safeguards around advanced AI models, while China is focusing on original innovation, moving away from its earlier “freerider” approach.
Also Read: Southeast Asia’s AI divide: SleekFlow report warns of widening gap
As Microsoft Vice Chair Brad Smith warned, “this race between the US and China for international influence likely will be won by the fastest first mover.”
Outlook: Innovation unchained, but at a cost
The AI sector represents a collision of hypercapitalism, global ambition, and creative destruction. While consumers benefit from “better, faster, cheaper” solutions and developers gain access to advanced tools, the path to profitability remains fraught with complexity.
The genie is out of the bottle, and the global monetisation race is only just beginning.
—
We’re building the most useful WA community for founders and enablers. Join here and be part of it.
The image was generated using ChatGPT.
Source: “Trends-Artificial Intelligence” by BOND
The post Burning billions: AI’s capital frenzy and its global implications appeared first on e27.