Microsoft CEO Satya Nadella has issued a warning to companies using AI models, cautioning that they are inadvertently sharing sensitive business data with model providers. In a blog post, Nadella argues that enterprises are paying twice for AI: once with money and again with proprietary knowledge. He explains that companies feeding data to AI models are essentially teaching them about their operations, which can be exploited by competitors. Nadella highlights that corrections made when models are wrong are distilled into institutional know-how, a type of knowledge that competitors cannot buy. This data exchange is a growing concern among AI industry leaders, including venture capitalists and enterprise software executives. Nadella argues that if AI companies can freely train on public data, enterprises should have the right to study and distill those models as well. He calls for companies to retain ownership of their data and build their own proprietary learning environments on the cloud. This approach would allow businesses to maintain control over their data and avoid dependency on a single AI provider. Source: techcrunch
Nadella’s concerns are echoed by industry players who are increasingly adopting open-source models. Idit Levine, CEO of Solo.io, reports that many enterprises are moving toward on-prem open-source models, which offer cost savings and greater control. She notes that these models can perform nearly 90% of the functions of proprietary models while being significantly cheaper. Companies like T-Mobile, ADP, and SAP are already using open-source models, and the trend is expected to grow. Tools like AI gateways and model-switching platforms are gaining popularity, with Vercel and OpenRouter reporting increased traffic to open-source models. Open-source models accounted for 29% of all traffic routed through Vercel’s gateway last month. Nadella’s call for data ownership aligns with this shift, as companies seek to avoid the risks associated with proprietary AI models. Source: techcrunch
Nadella’s warning is part of a broader debate about the ethical and strategic implications of AI model usage. He emphasizes that enterprises are not just consuming intelligence but also creating it, and that this created intelligence should belong to them. The CEO of Microsoft, which has invested in both OpenAI and Anthropic, is now openly urging companies to be cautious about using proprietary models. This stance reflects growing concerns within the industry about data privacy and model transparency. As more companies adopt open-source solutions, the balance of power between AI providers and their users is shifting. Source: techcrunch
