"State of play: Hyperscalers - the AI data-center giants - could spend up to $700 billion from their balance sheets on AI this year, while issuing eye-popping debt to access even more capital, according to UBS. The eight largest companies, all tech firms with AI ambitions, make up nearly half of the S&P 500. More than half of invested venture-capital dollars went to AI firms in 2025, per PitchBook."
"Zoom in: Anthropic recently raised $30 billion at a $380 billion post-money valuation, making it the world's fourth-most valuable private tech company. The Claude creator hit a $14 billion revenue run rate this month, growing 10x annually over the past three years. But CEO Dario Amodei has warned that sales may not keep up with the compute treadmill, as next-generation models could eventually cost up to $100 billion to train."
Hyperscalers could spend up to $700 billion from their balance sheets on AI this year while issuing substantial debt to access additional capital. The eight largest tech firms focused on AI comprise nearly half of the S&P 500 market capitalization, concentrating market exposure. More than half of venture-capital investments in 2025 flowed to AI firms, and private credit could supply half of the $1.5 trillion needed for data-center buildouts, shifting investment concentration into alternatives. Anthropic raised $30 billion at a $380 billion valuation and reached a $14 billion revenue run rate with rapid growth. Next-generation model training costs could reach $100 billion, creating a compute treadmill that may outpace monetization. AI stocks accounted for roughly 70% of the S&P 500 gains in 2025 and AI capex drove over 90% of economic growth in the first half of 2025, creating systemic market and economic risk if monetization falters.
Read at Axios
Unable to calculate read time
Collection
[
|
...
]