Lock-in builds up gradually: proprietary data formats, custom integrations built against one vendor's specific API, and years of accumulated configuration all raise the cost of leaving. AI adds a fast moving version of this risk, since prompts, fine-tuned models and integrations built tightly around one provider can be expensive to rebuild if that provider changes pricing or shuts down a model.
The common misconception is that avoiding lock-in means avoiding commitment to any vendor. In practice it means designing with a model-agnostic architecture and portable data formats from the start, so switching later is a real option and a negotiating position, even for a company with no near-term plan to switch.