Close Menu
    X (Twitter) LinkedIn
    CapitalAI DailyCapitalAI Daily
    X (Twitter) LinkedIn
    • Markets & Investments
    • Big Tech & AI
    • AI & Cybercrime
    • Jobs & AI
    • Banks
    • Crypto
    Sunday, February 8
    CapitalAI DailyCapitalAI Daily
    Home»Big Tech & AI»AI Intelligence Pricing Collapses From $37.50 to Pennies in a ‘Wicked’ Race to the Bottom, Says Marc Benioff

    AI Intelligence Pricing Collapses From $37.50 to Pennies in a ‘Wicked’ Race to the Bottom, Says Marc Benioff

    By Henry KanapiDecember 27, 20252 Mins Read
    Share
    Twitter LinkedIn

    Marc Benioff says artificial intelligence pricing is collapsing at an unprecedented pace, as leading models race toward near-zero costs per million tokens.

    In a new post on X, the Salesforce CEO points to dramatic price declines across the AI ecosystem, framing the shift as a structural reset rather than normal competition.

    Citing data from JPMorgan Asset Management, Benioff highlights how quickly costs have fallen from the early days of large language models, contrasting today’s prices with what developers were paying just two years ago.

    “GPT-4 in 2023: $37.50!”

    He says newer models now operate at a fraction of that cost, with prices measured in cents rather than dollars.

    “AI is on a wicked race to the bottom!

    Grok 4: $0.28/million tokens. Gemini 3 Flash: $0.50/million tokens. Gemini 2.5 Flash: $0.30/million tokens. Gemini 2.5 Flash-Lite: $0.10/million tokens.”

    Image
    Source: Marc Benioff/X

    According to Benioff, the collapse in pricing signals a fundamental change in how AI intelligence is produced and consumed, as efficiency gains and competition drive costs sharply lower.

    “Look out below, and let’s go!”

    His comments come as enterprises and developers increasingly treat AI intelligence as a commodity input, with falling prices accelerating adoption and reshaping the economics of software, services and automation.

    The data runs counter to the claim of “Big Short” investor Michael Burry, who said that LLMs are spending “tens of dollars” to answer a query. JPMorgan’s data shows that while Burry’s estimates may have been true two years ago, they no longer hold water. A typical response might use 1,000 to 5,000 tokens, costing far less than a dollar.

    Disclaimer: Opinions expressed at CapitalAI Daily are not investment advice. Investors should do their own due diligence before making any decisions involving securities, cryptocurrencies, or digital assets. Your transfers and trades are at your own risk, and any losses you may incur are your responsibility. CapitalAI Daily does not recommend the buying or selling of any assets, nor is CapitalAI Daily an investment advisor. See our Editorial Standards and Terms of Use.

    AI intelligence cost Marc Benioff News Tokens
    Previous ArticleRobert Kiyosaki Sees Potential 185% Rally for Precious Metal in 2026 As AI and Data Centers Drive New Demand
    Next Article FBI Warns AI Impersonation Is Undermining Trust at the Highest Levels of Government

    Read More

    Nvidia CEO Jensen Huang Says Nobody Uses AI Better Than This Company – And It’s Not Google or OpenAI

    February 7, 2026

    ‘We’re Totally Screwed’ – Elon Musk Issues Stark Warning on US Bankruptcy Risk

    February 6, 2026

    Here’s Why Amazon Commits to $200,000,000,000 Capital Spending This Year – ‘Those Are All Big Tailwinds’

    February 6, 2026

    Alibaba Founder Reveals the ‘Biggest Challenge’ in the Chinese AI Market Amid Reliance on Open-Source Models

    February 6, 2026

    ‘Magical Electricity Fairies?’ Elon Musk Warns AI Chips Growing Exponentially but May Go Unused

    February 6, 2026

    Google Shatters $400,000,000,000 Revenue Barrier As Gemini Drives Momentum, Says CEO Sundar Pichai

    February 5, 2026
    X (Twitter) LinkedIn
    • About
    • Author
    • Editorial Standards
    • Contact Us
    • Privacy Policy
    • Terms of Service
    • Cookie Policy
    © 2025 CapitalAI Daily. All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.