← Back to news

    OpenAI Backs OpenClaw’s Foundation Plan as Creator Peter Steinberger Joins the Company

    AI AgentsOpen SourceDeveloper Tools

    OpenClaw and OpenAI — foundation support

    Image: TechSwamy-generated illustration.

    BENGALURU (TechSwamy News) — OpenAI CEO Sam Altman said that Peter Steinberger, the creator of the viral open-source AI agent OpenClaw, is joining OpenAI, and that OpenClaw will “live in a foundation” as an open-source project that OpenAI will continue to support.

    The announcement is notable not just as a talent move, but as a signal: OpenAI is explicitly positioning multi-agent systems—agents that coordinate and take actions across tools—as a core part of what comes next.

    What OpenAI said

    In a public post on X, Altman said Steinberger is joining OpenAI “to drive the next generation of personal agents,” adding that the future is “extremely multi-agent,” and that OpenClaw will live on as an open-source foundation-backed project with OpenAI support.

    CNBC reported that no financial terms were disclosed.

    What Steinberger said: “open and independent”

    In his own note, Steinberger framed the move as a way to focus on building and shipping—without the overhead of running a standalone company—while keeping OpenClaw open.

    “I’m joining OpenAI to work on bringing agents to everyone. OpenClaw will move to a foundation and stay open and independent,” Steinberger wrote.

    He also wrote that OpenAI has made commitments to support his ability to dedicate time to the project, and that OpenAI already sponsors OpenClaw, while he works to place it into a foundation structure.

    Why a foundation model matters

    If executed well, a foundation structure can do two things at once:

    • Preserve open-source governance and contributor trust as the project scales.
    • Create durable resourcing (funding, security work, maintenance) for a tool that is increasingly used in production-like settings.

    For agentic systems—where users may grant access to email, calendars, browsers, and internal tools—sustained investment in safety and hardening is not optional.

    The larger implication: agents are moving from novelty to infrastructure

    OpenClaw’s breakout moment has been tied to a simple idea: move from “AI that answers” to AI that acts.

    Altman’s emphasis on multi-agent systems suggests OpenAI believes the next phase will involve:

    • agents coordinating with other agents,
    • tool-using workflows that span apps and services,
    • and stronger primitives for reliability and safety.

    Whether OpenClaw’s foundation becomes a major open ecosystem (or a feeder for proprietary agent products) will depend on how the structure and commitments translate into day-to-day governance. But the direction is clear: agent infrastructure is now strategic.

    Sources