{"slug": "the-substitution-wave-in-ai", "title": "The Substitution Wave in AI", "summary": "Three forces are reshaping AI cost structures as major buyers substitute cheaper models for expensive ones. Coinbase routed prompts to cheaper models to keep costs flat while token usage grew exponentially, Lindy saved millions by switching entirely to DeepSeek v4 from Anthropic, and Harvey found Kimi 2.6 achieved 11x cost savings over Opus on legal benchmarks. The substitution wave is driving buyers to reinvest savings into more intelligence rather than pocketing discounts, with open models becoming cheaper at parity while closed frontier models grow more expensive.", "body_md": "Three forces are reshaping the AI cost structure :\n\nThe natural response from AI buyers is substitution.\n\nCoinbase 6 :\n\nAt Coinbase we’re working hot on routing prompts to cheaper models where appropriate, & in some cases have been able to keep costs roughly flat, while token usage continues to grow exponentially.\n\nLindy 7 :\n\nPulled the trigger today & switched 100% of Lindy traffic to DeepSeek v4, churning from Anthropic models. Saves us millions of $ & we’re actually seeing an\n\nincreasein performance on many core use cases. Transformative for the business.\n\nHarvey 8 :\n\nOn a 100-task slice of our Legal Agent Benchmark (LAB), SFT moved Kimi 2.6’s all-pass rate from 11% to 15%, beating Opus’ 14%. But the cost gap was even more striking : $84 vs $954 across the same 100 tasks, or ~11x cheaper.\n\nCursor went further. They post-trained Kimi K2.5 into their own production model, Composer.[9](https://tomtunguz.com/index.xml#fn:9)\n\nComposer 2.5 is exceptionally intelligent & up to 10x more efficient than similarly capable models.\n\nCoinbase’s quote shows where the savings go : costs flat, tokens exponential. Buyers don’t pocket the discount — they spend it on more intelligence.\n\nClosed models are getting more expensive at the frontier; open models are getting cheaper at parity. The choice is which slope you want under your unit economics.", "url": "https://wpnews.pro/news/the-substitution-wave-in-ai", "canonical_source": "https://www.tomtunguz.com/inflation-deflation-ai/", "published_at": "2026-06-07 00:00:00+00:00", "updated_at": "2026-06-12 15:13:07.373345+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-startups", "ai-products", "ai-infrastructure"], "entities": ["Coinbase", "Lindy", "DeepSeek", "Anthropic", "Harvey", "Kimi", "Cursor", "Composer"], "alternates": {"html": "https://wpnews.pro/news/the-substitution-wave-in-ai", "markdown": "https://wpnews.pro/news/the-substitution-wave-in-ai.md", "text": "https://wpnews.pro/news/the-substitution-wave-in-ai.txt", "jsonld": "https://wpnews.pro/news/the-substitution-wave-in-ai.jsonld"}}