Goldman Sachs: AI's annual revenue gap hits $230B

Goldman Sachs estimates the five largest U.S. hyperscalers — Amazon, Alphabet, Microsoft, Oracle and Meta — need about $300 billion in annual AI revenue to break even on $800 billion in 2026 AI infrastructure spending. With AI cloud revenue running only about $70 billion above its pre-AI trend, the annual gap is roughly $230 billion.

• 2026 capex for the five hyperscalers is projected at $800B, up from $413B in 2025
• About $300B in annual AI revenue is needed to break even
• AI cloud revenue is only ~$70B above its pre-AI trend line
• Goldman projects 2027 AI capex at $1.2 trillion