Opus 4.8 Won Our Benchmark. I Still Wouldn't Use It For Everything.

For deeper playbooks and analysis: https://natesnewsletter.substack.com/What's really happening with Opus 4.8, Claude Code, and the AI model race in 2026?The common story is that a stronger model automatically becomes the default tool — but the reality is that harnesses, compute, reliability, and workflow design now matter just as much as raw model capability.In this episode, I share the inside scoop on why Opus 4.8 is a strong but complicated release, why it is not automatically my daily driver, and why Codex currently fits certain long-running agent workflows better.Why Opus 4.8 reads more like a checkpoint release than the Mythos moment people expectedHow reasoning effort can become unpredictable when a model overthinksWhat a harness is, and why it now decides daily-driver behaviorWhy Claude Code's /workflows command is a real agent-pattern innovationWhere knowledge workers and engineering leaders should focus in the second half of 2026This matters for builders, executives, CTOs, CIOs, and operators trying to decide where to place AI budget. The practical question is not which model wins forever. It is how you architect your work so you can route tasks to the model and harness that best drive the outcome.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information. Hosted on Acast. See acast.com/privacy for more information.