The promise of autonomous agentic AI requires significant changes in the governance landscape. Parents of young children face a lot of fears about developmental milestones, from infancy through adulthood. The number of months it takes a baby to learn to talk or walk is often used as a benchmark for wellness, or an indicator of additional tests needed to properly diagnose a potential health condition. A parent rejoices over the child's first steps and then realizes how much has changed when the child can quickly walk outside, instead of slowly crawling in a safe area inside. Suddenly safety, including childproofing, takes a completely different lens and approach. Generative AI hit toddlerhood between December 2025 and January 2026 with the introduction of no code tools from multiple vendors and the debut of OpenClaw, an open source personal agent posted on GitHub.

Y et, to our knowledge, every metric for alignment is a posteriori, i.e., a system is deemed aligned as

Among the research topics in multi-agent learning, mixed-motive cooperation is one of the most prominent challenges, primarily due to the mismatch between individual and collective goals.

Much of this effort has focused on the single-agent setting, in which an agent maximizes a predefined extrinsic reward function.

This supplementary document is structured as follows.

Autonomous agents will soon run thousands of enterprise workflows, and only organizations with unified, trusted, context-rich data will prevent chaos and unlock reliable value at scale. AI agents are moving beyond coding assistants and customer service chatbots into the operational core of the enterprise. The ROI is promising, but autonomy without alignment is a recipe for chaos. Business leaders need to lay the essential foundations now. Agents are independently handling end-to-end processes across lead generation, supply chain optimization, customer support, and financial reconciliation. A mid-sized organization could easily run 4,000 agents, each making decisions that affect revenue, compliance, and customer experience.

Human-AI shared control allows human to interact and collaborate with autonomous agents to accomplish control tasks in complex environments. Previous Reinforcement Learning (RL) methods attempted goal-conditioned designs to achieve human-controllable policies at the cost of redesigning the reward function and training paradigm. Inspired by the neuroscience approach to investigate the motor cortex in primates, we develop a simple yet effective frequency-based approach called Policy Dissection to align the intermediate representation of the learned neural controller with the kinematic attributes of the agent behavior. Without modifying the neural controller or retraining the model, the proposed approach can convert a given RL-trained policy into a human-controllable policy. We evaluate the proposed approach on many RL tasks such as autonomous driving and locomotion. The experiments show that human-AI shared control system achieved by Policy Dissection in driving task can substantially improve the performance and safety in unseen traffic scenes. With human in the inference loop, the locomotion robots also exhibit versatile controllable motion skills even though they are only trained to move forward. Our results suggest the promising direction of implementing human-AI shared autonomy through interpreting the learned representation of the autonomous agents. Code and demo videos are available at https://metadriverse.github.io/policydissect

The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recent LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress towards capable autonomous agents.