IBM's chief executive has highlighted a shift in the artificial intelligence landscape, noting that open weight models have reached a level of quality and affordability that makes them highly attractive for local deployment.
According to the CEO, these models are now performing well enough to be considered a viable option for businesses that prefer to keep their data and processing on-premises. The combination of high performance and lower costs for local hosting is creating a strong incentive for companies to move away from closed-source alternatives.
What it means
This perspective suggests that the competitive advantage previously held by proprietary AI providers is narrowing. As open weights become more accessible and efficient to run on internal infrastructure, the trade-off between convenience and control is shifting in favor of local deployments.
