The enterprise readiness gap that legal teams must navigate
Product counsel need to become infrastructure assessors, evaluating not just model capabilities but whether entire enterprise ecosystems can handle advanced AI before deployment.
Arun Chandrasekaran from Gartner captures where we are with AI right now: "All we have done is create some very good engines for a car, and we are getting super excited, as if we have this fully functional highway system in place." For product counsel, this mismatch between capability and infrastructure defines our work today.
GPT-5 shows clear improvements in coding tasks, multimodal capabilities, and tool orchestration. The model can reason better and hallucinates less, which makes it more suitable for enterprise use. But here's what I'm seeing in practice—enterprise systems that can't handle concurrent API requests, governance frameworks that weren't designed for 128K context windows, and oversight mechanisms built for simpler AI interactions.
Gartner found that agentic deployments still need narrow domains and human oversight, which undermines the autonomy organizations want. We have a window of opportunity where legal teams feel pressure to enable advanced AI capabilities without the supporting infrastructure being ready.
The answer isn't waiting for everything to be perfect. It's building legal frameworks that can evolve alongside model improvements while maintaining enterprise safety and compliance standards. This means product counsel needs to become infrastructure assessors—evaluating not just what the model can do, but whether the entire ecosystem can handle those capabilities.
I'm auditing identity management systems, validating data pipelines, and ensuring that governance policies can effectively handle the impact of these models before teams deploy them. The companies that coordinate technical advancement with legal and operational readiness will outperform those that let model capabilities outpace their infrastructure.
