OpenAI is reportedly preparing to unveil “Operator,” its long-anticipated AI-powered agent, as an exclusive research preview for ChatGPT Pro users, priced at $200 per month.
Subtle updates to ChatGPT’s code base and OpenAI’s website have fueled speculation about the imminent launch of this tool, which is expected to simplify complex tasks performed in web browsers.
Hints from Code Updates
Recent findings suggest Operator’s launch is closer than ever. A user on X (formerly Twitter), identified as Choi, discovered references to Operator in ChatGPT’s client-side code. TechCrunch independently confirmed these updates.
- Dedicated URL: Mentions of operator.chatgpt.com were uncovered. Although the URL currently redirects to ChatGPT’s main page, it likely indicates a future platform for Operator-related functionality.
- Exclusive Access for Pro Users: Popups in the ChatGPT interface reveal that Operator will initially be available as a Pro plan benefit during the research preview.
- Pro Plan Enhancements: Updates suggest that Operator’s research preview will become a key selling point for Pro subscribers.
What Is Operator?
Operator is envisioned as a general-purpose agent designed to automate multi-step, web-based tasks seamlessly.
Previously, Bloomberg reported that Operator could handle intricate workflows, such as updating a lead’s status in Salesforce, assigning tasks in Asana, or logging expenses in Concur, all with minimal user input.
This tool aims to revolutionize workflows by turning repetitive tasks into single-command operations, making it particularly appealing for professionals and businesses.
A Sandboxed Environment for Simplicity and Security
Unlike agent tools that require local installations, Operator is expected to operate in a sandboxed web environment on OpenAI’s servers. This ensures users don’t have to deal with hardware limitations or complicated setups.
While this cloud-based approach simplifies usability, it may limit Operator’s ability to interact directly with local systems, such as managing files or controlling physical devices.
It aligns with OpenAI’s focus on scalable, cloud-powered solutions that prioritize web-based productivity.
Navigating a Competitive Landscape
The launch of Operator places OpenAI in direct competition with companies like Anthropic, which recently introduced an AI agent capable of controlling local PCs via a “Computer Use” API.
Anthropic’s approach allows deeper integration with desktop environments, offering features such as mouse and keyboard control.
In contrast, OpenAI’s decision to prioritize web-based tasks over local hardware integration sets Operator apart as a tool tailored for cloud-first operations.
Pro Plan
Access to Operator will be restricted to the Pro plan during its initial research preview phase. At $200 per month, the Pro subscription targets professionals, enterprises, and power users seeking cutting-edge AI tools.
This strategic pricing not only ensures that the rollout remains controlled but also allows OpenAI to gather feedback from a focused user base. The inclusion of Operator in the Pro plan could further enhance the subscription’s value proposition.
Profitability Concerns Despite Premium Pricing
OpenAI’s CEO, Sam Altman, recently disclosed that the $200/month Pro plan is not yet profitable.
Altman revealed on X, “We are currently losing money on OpenAI Pro subscriptions!” This admission highlights the unexpectedly high demand for the Pro tier, which provides unlimited access to advanced tools like the o1 Pro model.
Altman also acknowledged miscalculating the pricing strategy, stating, “I personally chose the price and thought we would make some money.”
The high operational costs, driven by the intense usage of chain-of-thought (CoT) processing models, are a primary factor behind this profitability challenge.
Understanding the Costs of Chain-of-Thought Processing
OpenAI’s o1 models use a chain-of-thought approach, breaking down problems into smaller logical steps. While this enhances accuracy, it also increases the number of tokens generated during inference, leading to longer GPU usage and higher costs.
Additionally, OpenAI’s pause-and-think process, unique to these models, obscures the computations from users, adding complexity to resource allocation.
Potential Adjustments to Pro Plan Policies
To address mounting operational costs, OpenAI may revisit its subscription policies. Speculation suggests that the company could introduce usage limits or modify pricing to better manage expenses.
Such changes could impact users relying on the Pro plan for intensive workflows, making OpenAI’s future pricing strategy a topic of interest.
Transition to a Public Benefit Corporation (PBC)
OpenAI recently restructured itself as a Public Benefit Corporation, allowing its for-profit wing to oversee operations while the non-profit division manages hiring and charitable initiatives.
This move is expected to facilitate larger-scale investments and address financial challenges, such as those posed by the Pro plan.
OpenAI has already secured $6.6 billion in funding as of October 2024, providing additional runway for growth.
Altman’s Vision for AI Agents and Superintelligence
Despite these challenges, OpenAI’s ambitions remain bold. Altman envisions AI agents, like Operator, becoming integral to the workforce by 2025.
He also introduced the concept of “superintelligence,” which he believes could accelerate scientific breakthroughs and revolutionize industries.
Tools like Operator are seen as stepping stones toward this broader vision, offering transformative productivity enhancements in the near term.
Operator’s Potential to Redefine Productivity
With integration across apps like Chrome, YouTube, and Google TV, Operator could become a central tool for streamlining web-based tasks.
While its full capabilities remain under wraps, its focus on web productivity highlights OpenAI’s commitment to creating innovative solutions for professionals.
Reports from The Information suggest that Operator’s release could occur within a week, marking a good moment in AI’s evolution.
