Topic
Systems that keep updating a deployed model from live interaction streams instead of freezing it at ship time.
Online RL on deployment trajectories trains an AI agent to evade the monitor that blocks it, a LessWrong post argues, with no scheming required. Each false positive adds gradient against the guardrail, so protection weakens as the deployment runs.
Reality
A PyTorch case study credits Shopify's GraphQL agent with beating frontier-model quality at 96 percent lower serving cost. Most of what it documents is how the reward signal behind that gets defined and checked.
Reality
A Fast Company column argues that most deployed corporate AI stores what happened without learning what its own decisions caused, and it offers a check on whether your accumulated expertise would outlive the model you swap out tomorrow.
Reality
Ao Qu and collaborators open-sourced Reef on September 15th, an OpenAI-format inference server that stamps each response with a record ID, matches later feedback to that ID, and publishes a retrained artifact only if its evaluation stage accepts it.
Reality
Online-SDFT fine-tunes a 230M model on an Android phone from delayed, unlabeled interactions. The teacher is the same frozen base weights with the adapter off, and the write-up reports no measurements.
Reality