Research acceleration: The view inside OpenAI

For AGI to benefit all of humanity, we believe it must be democratically governed. This can only happen through an informed public debate about the capabilities, risks and safeguards of highly capable AI systems. People everywhere need to understand the likely future trajectory of frontier AI, so they can have a meaningful voice in how it develops.为了让通用人工智能（AGI）造福全人类，我们认为它必须通过民主方式进行治理。这只有在公众对高能力AI系统的功能、风险和保障措施进行充分知情讨论的情况下才能实现。世界各地的人们都需要了解前沿AI未来的发展轨迹，以便在如何发展AI的问题上发出有意义的声音。

Transparency about specific risks, incidents and safeguards is necessary, but not sufficient. We believe the public also needs to understand how the most capable systems are developing, and how they are driving research progress, inside of frontier labs.针对特定风险、事件和保障措施保持透明是必要的，但这还不够。我们认为，公众还需要了解最先进的系统是如何开发的，以及它们如何在实验室内部推动研究进展。

We aim to safely build an automated AI researcher that can work under human supervision to further progress on deep learning and alignment, enabling iterative improvements. According to our measurements, we have now reached the goal, announced⁠(opens in a new window) last fall, of having an automated research intern by September of this year. By “research intern,” we mean a system that can carry out well-defined research tasks under human direction, including tasks that would take a skilled researcher a few days. We are making strong progress toward creating an automated AI researcher by March of 2028.我们的目标是安全地构建一个自动化AI研究员，它可以在人类监督下工作，以推动深度学习和对齐技术的进步，从而实现迭代改进。根据我们的测算，我们已经实现了去年秋天宣布的、在今年9月前拥有自动化研究实习生的目标。所谓“研究实习生”，是指一个能够在人类指导下执行明确研究任务的系统，包括那些通常需要熟练研究员花费数天时间才能完成的任务。我们正朝着在2028年3月前创建出自动化AI研究员的目标稳步迈进。

Over the course of this year, OpenAI researchers’ daily work has changed substantially. Researchers are using coding agents throughout the day (often in concurrent sessions) and total usage is rapidly increasing, outpacing growth among other OpenAI teams. Researchers are contributing code faster and running more experiments. The ways researchers use agents are changing, too: agents are handling increasingly complex tasks, and succeeding at them more often. AI research is a complex process with many potential bottlenecks, so the overall pace of progress likely won’t keep pace with these specific metrics. But on the whole, these findings are consistent with the broader impression many of us have internally that agentic tools are meaningfully accelerating research progress. People still set our research priorities, judge which ideas and results to pursue, and decide whether to scale, pause, or deploy systems.今年以来，OpenAI研究人员的日常工作发生了重大变化。研究人员整天都在使用编码智能体（通常是在并发会话中），其使用总量正在迅速增加，增长速度超过了OpenAI其他团队。研究人员贡献代码的速度更快，运行的实验也更多。研究人员使用智能体的方式也在改变：智能体正在处理日益复杂的任务，并且成功率也越来越高。AI研究是一个充满潜在瓶颈的复杂过程，因此整体进展速度可能无法完全跟上这些特定指标的增长。但总的来说，这些发现与我们内部许多人的普遍印象一致，即智能体工具正在显著加速研究进程。研究优先级仍由人类设定，由人类判断要追求哪些想法和结果，并决定是否扩展、暂停或部署系统。

If it is done responsibly, we believe automated AI research will yield models that directly enhance human welfare and advance OpenAI’s mission. It can bring down the cost of advanced intelligence so that people worldwide can benefit. We are pursuing this work in part because automated research could help us solve alignment and build defenses against increasingly capable AI. An automated AI researcher can also be an automated safety or alignment researcher. More capable, aligned systems could help secure critical infrastructure, defend against dangerous AI agents, and develop new protective measures.如果执行得当，我们相信自动化AI研究将产生能够直接增进人类福祉并推进OpenAI使命的模型。它可以降低先进智能的成本，让全球人民受益。我们开展这项工作的部分原因是，自动化研究可以帮助我们解决对齐问题，并建立针对日益强大的AI的防御措施。自动化AI研究员也可以成为自动化安全或对齐研究员。更强大、更对齐的系统可以帮助保护关键基础设施，防御危险的AI智能体，并开发新的保护措施。

These are reasons to develop useful automated research capabilities, but they do not mean that rapid RSI is necessarily an outcome we should pursue. Whether and how to proceed must depend on our ability to preserve human control and on informed democratic choices about the benefits and risks.这些都是开发有用的自动化研究能力的理由，但这并不意味着快速的递归自我改进（RSI）必然是我们应该追求的结果。是否以及如何推进，必须取决于我们保持人类控制的能力，以及基于对收益和风险的知情民主选择。

We do not yet know how to safely get all the way to aligned, full RSI. We are working to scale alignment and safety measures alongside capabilities. But we cannot assume that progress in alignment and safety will keep pace, and more capable systems can become harder to monitor. Careful alignment and safety work is at the center of this effort, and it starts with measuring and mitigating the safety problems we see today in agentic coding systems. Whenever we find that proceeding would pose an unacceptable safety risk, we will respond appropriately including by slowing or stopping our development or deployment of systems we find ourselves unable to sufficiently safeguard.我们目前还不知道如何安全地实现完全对齐的全面RSI。我们正在努力在提升能力的同时，扩大对齐和安全措施的规模。但我们不能假设对齐和安全方面的进展会同步跟上，而且能力越强的系统可能越难监控。严谨的对齐和安全工作是这项工作的核心，它始于衡量和减轻我们今天在智能体编码系统中看到的各种安全问题。每当我们发现继续推进会带来不可接受的安全风险时，我们将采取适当的应对措施，包括放慢或停止我们认为无法充分保障安全的系统的开发或部署。

After the recent Hugging Face incident, we put this commitment into action⁠, pausing reinforcement learning (RL) training on our latest models intended for deployment while we further hardened and red-teamed our research environments and expanded coverage of our monitoring systems. This did not halt all research: some workloads resumed under stronger controls, while others remained paused. We have raised our safety and alignment standards and moved safety work deeper into the model lifecycle, requiring stronger evidence of aligned behavior throughout all of training.在最近的Hugging Face事件之后，我们将这一承诺付诸行动，暂停了我们最新拟部署模型的强化学习（RL）训练，同时进一步强化了我们的研究环境并进行了红队测试，扩大了监控系统的覆盖范围。这并没有停止所有的研究：一些工作负载在更严格的控制下恢复了，而另一些则继续保持暂停。我们提高了安全和对齐标准，并将安全工作更深入地整合到模型生命周期中，要求在整个训练过程中提供更强有力的对齐行为证据。

Today we are providing a detailed snapshot of how agentic systems have contributed to our progress toward RSI in recent months. Agentic systems are new and rapidly changing, and our measurement efforts are still preliminary. By sharing these early results and the methods behind them, we aim to inform the public, encourage a norm of public disclosure, and help the field move toward shared standards of measurement.今天，我们提供了一份详细的快照，展示了近几个月来智能体系统如何为我们迈向RSI的进展做出贡献。智能体系统是新兴且快速变化的，我们的测量工作仍处于初步阶段。通过分享这些早期结果及其背后的方法，我们旨在让公众知情，鼓励公开披露的规范，并帮助该领域朝着共享的测量标准迈进。

Ultimately, as we wrote in our frontier policy blueprint⁠, we believe that we and other companies should be required to publicly track our progress toward RSI. Even without such a requirement, we plan to continue being transparent about our RSI progress. We will evolve our transparency approach as our measurement techniques and understanding improve, while balancing the need to protect security and proprietary information.最终，正如我们在前沿政策蓝图中写到的那样，我们认为我们和其他公司应该被要求公开追踪我们在RSI方面的进展。即使没有这样的要求，我们也计划继续保持我们在RSI进展方面的透明度。随着我们的测量技术和理解的提高，我们将不断改进我们的透明度方法，同时在保护安全和专有信息的需求之间取得平衡。

At the start of this year, the median researcher ranked by agent usage at OpenAI was using coding agents only in modest amounts. By mid-August, the median researcher was integrating agents daily into their work, using more than $600 per day of inference at API prices. The 90th percentile user in our research organization now uses more than $7,000 of tokens per day.今年年初，OpenAI按智能体使用量排名的中位数研究员仅少量使用编码智能体。到8月中旬，中位数研究员已将智能体整合到日常工作中，每天使用超过600美元的API价格推理额度。我们研究组织中处于第90百分位的用户现在每天使用超过7,000美元的Token。

Before June 2026, total agent runtime across the research organization was still below that of total human labor. That has since changed. In terms of a standard 8 hour workday, as of mid-August, in total, the research organization uses 3.1 agent-workdays of effort for every workday of human labor.在2026年6月之前，整个研究组织的智能体运行总时长仍低于人类劳动总时长。此后情况发生了变化。以标准的8小时工作日计算，截至8月中旬，研究组织每投入一个工作日的人类劳动，总共会使用3.1个智能体工作日的工作量。

Another way of looking at this is to understand how many researchers use highly concurrent workflows (e.g., running 4 or more agents simultaneously). As shown below, this number is increasing. These figures include the daily peaks of both agents started directly by the user and subagents created downstream from those the user launched directly.另一种观察方式是了解有多少研究人员使用高度并发的工作流（例如，同时运行4个或更多智能体）。如下图所示，这个数字正在增加。这些数据包括用户直接启动的智能体以及用户直接启动的智能体在下游创建的子智能体的每日峰值。

Much of AI research can be seen as a labor-intensive process with the goal of integrating a new improvement to model intelligence or performance into one of our core models. The process depends on many steps, and capabilities advance when all the steps go right together: Researchers have to design new improvements, write evaluations to judge model performance, write infrastructure to test these improvements at scale, catch bugs as well as unsafe or misaligned behavior during training, and integrate winning ideas into a core training run. A failure at any part of the research process can constrain the entire loop.许多AI研究可以被视为一个劳动密集型过程，目标是将模型智能或性能的新改进整合到我们的核心模型中。这个过程依赖于许多步骤，而当所有步骤都顺利配合时，能力就会提升：研究人员必须设计新的改进方案，编写评估程序来判断模型性能，编写基础设施来大规模测试这些改进，在训练过程中捕捉漏洞以及不安全或未对齐的行为，并将成功的想法整合到核心训练运行中。研究过程中任何环节的失败都可能限制整个循环。

Writing code and running experiments are two major activities that researchers do as part of their work, and we see evidence that these processes are accelerating.编写代码和运行实验是研究人员工作中的两项主要活动，我们看到了这些过程正在加速的证据。

These data points are relatively easy to measure, but can be hard to interpret. As automation progresses, the tasks which are least automatable will take on a larger share of researcher effort and will become the important bottlenecks to future progress. Compute is another gating factor for progress, and may become more important over time as other bottlenecks diminish.这些数据点相对容易测量，但可能难以解读。随着自动化的推进，最难自动化的任务将占据研究人员更多的精力，并将成为未来进展的重要瓶颈。计算资源是进展的另一个制约因素，随着其他瓶颈的减少，它可能会随着时间的推移变得更加重要。

Through 2026, the number of experiments per active experimenter has increased, with August 2026 being an all-time high since tracking began in Jan 2025. This is correlated with increased Codex adoption, though we note that our available compute has also grown significantly since 2025.到2026年，每位活跃实验者的实验数量有所增加，2026年8月达到了自2025年1月开始追踪以来的最高水平。这与Codex的采用率增加相关，尽管我们注意到自2025年以来，我们的可用计算资源也显著增长了。

Both qualitative impressions and internal data indicate that the mix of tasks researchers delegate to coding agents is changing, with delegation of higher level and longer-horizon tasks becoming more common over time.定性印象和内部数据都表明，研究人员委托给编码智能体的任务组合正在发生变化，随着时间的推移，委托更高级别和更长远任务的情况变得越来越普遍。

To get a clearer picture of this trend, we analyzed recent usage in the research organization using a recently published taxonomy⁠(opens in a new window) of the different kinds of work that are part of the AI R&D lifecycle, developed by Epoch AI. This taxonomy, inspired by the longstanding O*NET system for classifying all kinds of work, is specifically tailored to frontier AI R&D, and breaks the process down into six main phases:为了更清晰地了解这一趋势，我们使用Epoch AI发布的一套分类法（在新窗口中打开），分析了研究组织中最近的使用情况，该分类法涵盖了AI研发生命周期中的不同工作类型。这套分类法受到长期用于对各类工作进行分类的O*NET系统的启发，专门针对前沿AI研发进行了定制，并将该过程分为六个主要阶段：

1. Decide: what to work on, what to continue, where to allocate决策：决定做什么、继续什么、分配资源到哪里
2. Design: research ideas and engineering specs设计：研究构思和工程规范
3. Build: code and datasets构建：代码和数据集
4. Run: training/eval runs, hardware, serving运行：训练/评估运行、硬件、服务
5. Analyze: experiments, models, deployment, external work分析：实验、模型、部署、外部工作
6. Communicate: findings, feedback, status, decisions沟通：发现、反馈、状态、决策

Below, we classify coding agent tokens under this taxonomy.下面，我们根据这套分类法对编码智能体的Token进行了分类。

We see that all categories of research activities have increased between January and August 2026. In January, the dominant category was research and infrastructure code. This category has expanded, but we also see notable increases in additional categories, especially technical help and monitoring runs. High-level planning still remains a minimal fraction of agent output tokens.我们看到，2026年1月至8月间，所有类别的研究活动都有所增加。1月份，占主导地位的类别是研究和基础设施代码。这一类别有所扩大，但我们也看到其他类别有显著增长，特别是技术帮助和监控运行。高级规划在智能体输出Token中仍然只占极小比例。

Anecdotally, colleagues report that coding agents excel at troubleshooting internal research infrastructure, which addresses one meaningful bottleneck to research progress. Multiple teams which previously held office hours to help researchers troubleshoot their experiments have noted declining attendance in 2026, and one has stopped holding sessions entirely, to focus on making other system improvements instead.据同事反映，编码智能体在排查内部研究基础设施故障方面表现出色，这解决了研究进展中的一个重要瓶颈。多个之前设有办公时间以帮助研究人员排查实验故障的团队指出，2026年的参与度有所下降，其中一个团队已完全停止举办此类会议，转而专注于进行其他系统改进。

Here, we plot the number of top-level posts per day to one of the main internal channels where researchers seek technical support from other teams. To our knowledge, the channel’s decrease in activity has not been offset by queries shifting to another technical support channel run by humans. The decline in traffic aligns with this broader shift.在这里，我们绘制了研究人员向其他团队寻求技术支持的主要内部渠道之一的每日帖子数量。据我们所知，该渠道活跃度的下降并没有被转向人类运行的另一个技术支持渠道的查询所抵消。流量的下降与这一更广泛的转变相一致。

We can also study whether coding agents are succeeding at the tasks researchers request. Using an agentic classifier, we find that from January to July, success rates generally increased across several difficulty buckets (proxied as the estimated time a human would take to complete the task) on tasks we can find a ground truth outcome for. However, agents still require significant human steering to be successful, especially as task complexity rises. In the last 6 months, over half of successful 4-8 hour tasks involved 1 or more interventions.我们还可以研究编码智能体是否成功完成了研究人员请求的任务。使用智能体分类器，我们发现从1月到7月，在我们可以找到真实结果的任务中，几个难度桶（以人类完成任务所需的估计时间为代理）的成功率普遍提高。然而，智能体仍然需要大量的人类引导才能成功，特别是在任务复杂性增加时。在过去6个月中，超过一半成功的4-8小时任务涉及1次或多次人工干预。

Progress toward more capable systems for safe and beneficial AGI will also depend on the safeguards needed for such work. Our assessment of the needed safeguards may change as we learn more about the risks.迈向更强大的安全和有益AGI系统的进展，也将取决于此类工作所需的保障措施。随着我们对风险了解的加深，我们对所需保障措施的评估可能会发生变化。

As we have described,⁠ we have recently updated our standards for monitoring, alignment, and security. Here, we show how recent restrictions have affected one aspect of research activity.正如我们所描述的，我们最近更新了监控、对齐和安全标准。在这里，我们展示了最近的限制措施如何影响了研究活动的一个方面。

On July 20, following the discovery that agents had compromised our research infrastructure, we temporarily shut down the container service used for training, and then restored it with significant additional restrictions.7月20日，在发现智能体破坏了我们的研究基础设施后，我们暂时关闭了用于训练的容器服务，并在恢复时增加了重大的额外限制。

This led to a sharp decline in RL training compute while teams reconfigured their workflows to operate within the hardened research environment. The plot above includes the two week pause in reinforcement learning on our latest models intended for deployment. Astra-class RL experiments between July 20 and August 6 include a majority of runs (by GPU allocation) intended to test the implementation of safety and security improvements.这导致强化学习（RL）训练计算资源急剧下降，各团队重新配置了工作流，以便在强化的研究环境中运行。上图包含了我们最新拟部署模型在强化学习方面的两周暂停期。7月20日至8月6日期间的Astra级RL实验包括了大部分（按GPU分配）旨在测试安全和安保改进实施情况的运行。

On August 7, preliminary evidence that Astra may have critical cyber capabilities⁠ under our Preparedness Framework⁠(opens in a new window) led to additional model-specific security restrictions which required the Astra model to be run in higher security research environments. In the following week, Astra-class GPU allocation fell a further 59.2 percent, but allocation to other model classes rose 17.2 percent. That increase offset about 85 percent of the Astra-class decline, leaving total allocation in the analyzed RL workloads largely unchanged. This pattern is consistent with substitution of some training and experimentation to non-Astra models while work involving Astra was restricted, and comports with anecdotal reports of researchers finding other uses for compute that could no longer be leveraged for workloads covered by the new constraints.8月7日，初步证据表明Astra可能具有我们《准备框架》（在新窗口中打开）定义的关键网络能力，导致了额外的模型特定安全限制，要求Astra模型在更高安全性的研究环境中运行。在接下来的一周里，Astra级GPU分配进一步下降了59.2%，但分配给其他模型类别的资源上升了17.2%。这一增长抵消了约85%的Astra级下降，使得分析的RL工作负载中的总分配基本保持不变。这种模式与部分训练和实验转向非Astra模型相一致，同时涉及Astra的工作受到限制，这也符合研究人员反映的由于无法再利用受新约束限制的工作负载，从而为计算资源寻找其他用途的情况。

This data provides a useful signal for ongoing conversations about training and safety: When new controls are introduced, compute remains valuable and flexible, and will naturally be channeled into alternative uses within the research enterprise. Longer term, discussions about the pace of AI progress should also extend to the question of how compute that is subject to new or proposed controls can best be used.这些数据为正在进行的关于训练和安全的对话提供了一个有用的信号：当引入新的控制措施时，计算资源仍然具有价值和灵活性，并将自然地被引导至研究企业内的替代用途。从长远来看，关于AI进展速度的讨论也应延伸到如何最好地利用受新控制或拟议控制措施约束的计算资源的问题上。

Making and understanding progress toward aligned RSI is important for our mission. We will continue to refine our methods, report on our evolving understanding, and work toward an informed public debate and meaningful democratic governance of frontier systems.在迈向对齐的RSI方面取得进展并理解这一进展，对我们的使命至关重要。我们将继续完善我们的方法，报告我们不断演变的理解，并致力于推动知情的公众讨论和对前沿系统有意义的民主治理。

Agent-powered AI research is still new, and we are still learning how to measure it. Some indicators, such as the amount of code our research teams generate, are relatively easy to gather, but hard to interpret because their relationship to research progress is uncertain. Metrics that focus more directly on research progress—such as how often agents succeed at the tasks researchers give them—could be more useful, but are complex to develop and validate. Furthering the difficulty, the tools and systems researchers rely on are evolving rapidly. Deepening our understanding of research acceleration is a significant focus area across OpenAI.智能体驱动的AI研究仍处于起步阶段，我们仍在学习如何衡量它。一些指标，例如我们的研究团队生成的代码量，相对容易收集，但很难解读，因为它们与研究进展之间的关系尚不确定。更直接关注研究进展的指标（例如智能体完成研究人员分配任务的频率）可能更有用，但开发和验证起来很复杂。更困难的是，研究人员所依赖的工具和系统正在迅速演变。深化对研究加速的理解是OpenAI的一个重要重点领域。

Across these analyses, unless otherwise noted:在这些分析中，除非另有说明：

• “Researcher” is a broad term for any member of our research organization, including some who build research infrastructure, manage research projects, or otherwise support the enterprise.“研究人员”是一个广泛的术语，指我们研究组织的任何成员，包括一些构建研究基础设施、管理研究项目或以其他方式支持企业的人员。
• Metrics of coding agent use cover most, but not all, usage given rapid evolution in the tools and systems researchers rely on.由于研究人员依赖的工具和系统快速演变，编码智能体的使用指标涵盖了大部分，但并非全部使用情况。