American companies mention open-source model times up about 6-fold year over year
According to a Financial Times investigation, open-weight models are spreading from technology companies to traditional U.S. enterprises. AlphaSense data show that this year, in August and September, the number of times U.S. companies mentioned open-weight or open-source models during earnings calls and investor meetings increased by about 6 times year over year. PNC Financial, logistics firm CH Robinson, and Siemens have recently publicly discussed related applications. Tinder’s AI annualized spending rose from about $1 million in January to about $10 million in July; it has now begun routing some ordinary requests to open-weight models. AT&T says that, currently, about 40% of its AI tasks are handled by open models and it plans to increase that to about 70%. Its internal AI systems process about 45 billion tokens per day; for simple tasks it uses lower-cost models, while for complex tasks it continues to use cutting-edge models such as OpenAI and Anthropic.
According to a Financial Times investigation, open-weight models are spreading from technology companies to traditional U.S. enterprises. AlphaSense data show that this year, in August and September, the number of times U.S. companies mentioned open-weight or open-source models during earnings calls and investor meetings increased by about 6 times year over year. PNC Financial, logistics firm CH Robinson, and Siemens have recently publicly discussed related applications. Tinder’s AI annualized spending rose from about $1 million in January to about $10 million in July; it has now begun routing some ordinary requests to open-weight models. AT&T says that, currently, about 40% of its AI tasks are handled by open models and it plans to increase that to about 70%. Its internal AI systems process about 45 billion tokens per day; for simple tasks it uses lower-cost models, while for complex tasks it continues to use cutting-edge models such as OpenAI and Anthropic.