The stock market is currently deeply divided. On the one hand, companies whose business models are positively associated with artificial intelligence (AI) are experiencing an unprecedented rise in valuations. On the other hand, highly profitable, high-quality companies are being ignored by investors\u2014or radically punished\u2014despite record results in revenue, profits, and cash flow, driven by fear of a supposed \u201cSaaSpocalypse.\u201d<\/p>\n
This dynamic has led to the beneficiaries of the boom gaining ever greater influence in the capital markets, with just a few tech giants\u2014representing a historically high concentration\u2014now dominating the major stock market indices. The market capitalization of the ten largest companies currently accounts for about 37% of the S&P 500. The influx of funds via passive ETF products has further fueled this trend.<\/p>\n
Expectations for the AI boom currently seem to know no bounds: Analysts are forecasting a historic record volume of several hundred billion U.S. dollars for 2026, which is expected to be spent solely by the leading \u201chyperscalers\u201d\u2014Microsoft, Alphabet, Amazon, Meta, and Oracle\u2014as capital expenditures (Capex) for the expansion of data centers.<\/p>\n
The ongoing boom is ignoring a fundamental transformation taking place within these companies: They are evolving at record speed from asset-light business models with high free cash flows and returns on capital into capital-intensive heavyweights whose operating cash flows are no longer sufficient to cover their massive spending. Hyperscalers are increasingly emerging as major issuers in the bond market. According to Goldman Sachs, they have raised approximately $200 billion in new debt this year alone. This represents about 9% of all investment-grade bond supply this year.<\/p>\n
The core dilemma for investors is the mismatch between immediate capital outlays and uncertain future returns (ROI). Furthermore, the current depreciation schedules for expensive AI infrastructure are coming under scrutiny. While industry experts at Princeton University estimate the economic useful life of chips to be just 1 to 3 years due to technological obsolescence and severe thermal and power stress, many hyperscalers are depreciating these investments over a much longer period of 5 to 6 years. Should economic realities force a shortening of these schedules, massive write-downs on remaining book values loom, potentially severely eroding the profitability of AI investments.<\/p>\n
To date, the AI sector operates primarily within a closed B2B investment loop, with tech companies essentially buying hardware from one another. End-user monetization at the application layer remains miles behind the massive upfront capital expenditures. Furthermore, powerful Large Language Models (LLMs) are increasingly being commoditized, standardized, and offered via low-cost open-source alternatives. Consequently, it remains highly questionable whether these staggering investments will ever yield the anticipated returns.<\/p>\n
Parallels to previous technology cycles in economic history are visible. The expansion of the U.S. railroad network in the 19th century, the laying of the first telephone lines, and the fiber-optic build-out of the late 1990s all shared a seemingly insatiable appetite for infrastructure capital. In every instance, once the euphoria faded, the market came to the painful realization that actual end-user demand and true monetization lagged significantly behind. Overcapacity, collapsing prices, industry consolidation, and bankruptcies were the inevitable result.<\/p>\n
In the fixed-income market, we therefore maintain a risk-off position, remaining focused on the stability and predictability of coupon payments. In equities, we avoid companies with eroding free-cash-flow yields. Instead, our focus rests on established, highly profitable quality enterprises trading at attractive valuations well away from the AI hype.<\/p>\n","protected":false},"excerpt":{"rendered":"
The stock market is currently deeply divided. On the one hand, companies whose business models are positively associated with artificial intelligence (AI) are experiencing an unprecedented rise in valuations. On the other hand, highly profitable, high-quality companies are being ignored by investors\u2014or radically punished\u2014despite record results in revenue, profits, and cash flow, driven by fear […]<\/p>\n","protected":false},"author":7,"featured_media":4122,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","inline_featured_image":false,"mc4wp_mailchimp_campaign":[],"footnotes":""},"categories":[60],"tags":[],"class_list":["post-15996","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-gane-express-en"],"acf":[],"yoast_head":"\n