{
  "id": 1524033,
  "title": "AI Labs Stop Competing on Smarts and Start Competing on Price",
  "url": "https://urgent.news/2026/08/17/ai-labs-stop-competing-on-smarts-and-start-competing-on-price",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-17T17:07:12.000Z",
  "source": {
    "name": "PYMNTS",
    "slug": "pymnts",
    "url": "https://www.pymnts.com/news/artificial-intelligence/2026/ai-labs-stop-competing-on-smarts-and-start-competing-on-price/"
  },
  "original_language": "en",
  "account": "By October 2024, the cost of querying AI models at GPT-3.5's level had dropped from $20 per million tokens to $0.07 per million tokens, a staggering over-280-fold reduction in just two years, according to Stanford's AI Index 2025 Report. Companies price their models based on the million tokens processed. In the past six weeks, OpenAI, Google, and Anthropic have all slashed or halted these prices. Andreessen Horowitz, a venture capital firm, has been tracking this trend since 2024, referring to it as \"LLMflation,\" a roughly 10-fold price decline each year for a model of fixed capability. The pace accelerated this summer as the three largest AI labs simultaneously reduced their prices. OpenAI cut the cost of its fastest and cheapest model, GPT-5.6 Luna, by 80%, from $1 to $0.20 per million input tokens. It also reduced its mid-tier model, GPT-5.6 Terra, by 20%, from $2.50 to $2. Google cut the price of its Gemini Flash model, introduced on July 21, from $1.50 to an introductory $0.75 per million input tokens within weeks. It also priced its newest model, Gemini 3.7 Flash, at the same introductory $0.75 per million input tokens through the end of the year. Anthropic went a step further by reversing a price increase it had previously announced, making the $2 rate permanent. This trend mirrors the history of broadband internet, where the cost of a basic connection fell over time, enabling widespread adoption. As AI prices continue to drop across all tiers, even the most costly models, more enterprises are able to leverage AI for tasks that were previously too expensive.",
  "summary": "The cost of querying an artificial intelligence (AI) model performing at GPT-3.5’s level fell from $20 per million tokens in November 2022 to $0.07 per million tokens by October 2024, an over-280-fold reduction in about two years, according to Stanford’s AI Index 2025 Report. Tokens are the small chunks of text an AI model reads […] The post AI Labs Stop Competing on Smarts and Start Competing on…",
  "key_points": [
    "AI model costs dropped 280-fold in two years to $0.07 per million tokens",
    "OpenAI, Google, Anthropic slashed or halted prices in past six weeks",
    "Trend mirrors broadband internet cost reduction enabling widespread adoption"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}