Expired60% confidenceFactExact time
A 2024 Goldman Sachs research report estimated that AI infrastructure construction has already cost over $100 billion, yet the direct revenue contribution attributable to AI remains very limited.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/5/2026(expired)
Sources
AI泡沫警报：银行与云巨头为何开始集体示警
hackernewshackernews7/6/2026
Related Claims
VerifiedInference, rather than training, is becoming the primary source of AI infrastructure costs.75% similarUnverifiedCurrent AI development exhibits two main threads: stacking compute and infrastructure through massive funding, and enhancing model application value by competing for scarce specialized data.73% similarVerifiedInference cost, deployment efficiency, and multi-model support capability are becoming the core competencies of AI infrastructure-layer companies.71% similarUnverifiedBy fine-tuning existing open-source models, AI development costs can drop sharply from the tens-of-millions range to tens of thousands of yuan or even lower.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/196863API
curl https://kongchang.com/api/v1/knowledge/claims/196863MCP
get_claim(id=196863)