Google DeepMind CEO Demis Hassabis Questions OpenAI's Early Move into ChatGPT Ads
Demis Hassabis expresses surprise at OpenAI's decision to test ads in ChatGPT, highlighting concerns over user trust and the role of assistants in AI monetization.

The financial markets witnessed a historic realignment on Monday as the traditional insurance brokerage sector faced its most significant single-day decline in over a decade. The catalyst for this market turmoil was neither a catastrophic natural disaster nor a regulatory crackdown, but a software update. OpenAI, the company that ignited the generative AI revolution, officially approved and launched the first fully autonomous insurance application on its ChatGPT platform.
Dubbed "SureAgent," this new plugin leverages OpenAI’s advanced reasoning models to bypass traditional intermediaries entirely, allowing users to assess risk, compare complex policies, and bind coverage in seconds. The immediate reaction from Wall Street was swift and brutal. Investors, sensing an existential threat to the commission-based business models that have dominated the industry for centuries, triggered a massive selloff in major insurance broker stocks.
For decades, insurance brokers have served as the essential gatekeepers of the risk industry, navigating the labyrinth of policy fine print for corporate and individual clients. The arrival of an AI agent capable of performing this nuanced advisory role with greater speed and zero commission fees has raised a fundamental question: Is the era of the human insurance middleman drawing to a close?
The selloff began shortly after the opening bell, coinciding with the official press release from OpenAI and the live demonstration of the app. By midday, trading was halted for several mid-cap brokerage firms due to volatility. The sentiment was clear: the market views this technological leap not merely as a new competitor, but as a potential replacement for the brokerage value chain.
The hardest hit were firms specializing in personal lines and small-to-medium enterprise (SME) commercial insurance—sectors where the new AI application claims to offer immediate, fully underwritten quotes.
| Ticker | Company Name | Sector Focus | Daily Change (%) |
|---|---|---|---|
| AJG | Arthur J. Gallagher & Co. | Commercial/Risk Mgmt | -12.8% |
| AON | Aon plc | Global Professional Services | -9.4% |
| MMC | Marsh & McLennan | Risk Strategy | -8.1% |
| WLTW | Willis Towers Watson | Advisory/Broking | -7.6% |
| BROWN | Brown & Brown, Inc. | Insurance/Reinsurance | -10.2% |
Data reflects market close on February 9, 2026.
Analysts at Morgan Stanley downgraded the entire sector to "Underweight" in a note released late Monday, citing "structural obsolescence risks." The note highlighted that while complex, multinational risk portfolios might still require human intervention, the "low-hanging fruit" that constitutes the bulk of broker revenue is now directly in the crosshairs of autonomous AI agents.
The disruptive power of the new ChatGPT integration lies in its ability to synthesize vast amounts of actuarial data and policy language instantly. Unlike previous generations of "InsurTech" comparators, which simply aggregated prices, SureAgent acts as a fiduciary advisor.
According to technical documentation released by the developer—a stealth startup backed by OpenAI’s startup fund—the system utilizes a specialized "Reasoning Engine." This engine performs three critical functions previously thought to require human judgment:
Sam Altman, CEO of OpenAI, addressed the launch in an interview with CNBC, framing the development as a natural evolution of the platform’s capabilities. "We are moving from a world where AI retrieves information to a world where AI executes complex tasks," Altman stated. "Insurance is a prime example of an industry where the friction costs—commissions, paperwork, time—are artificially high. AI brings those costs down to near zero."
The primary driver of the stock panic is the threat to the commission model. Traditional brokers typically earn between 10% and 15% of the premium paid by the customer. SureAgent, by contrast, charges a flat monthly subscription fee of $20 for unlimited access to its advisory services, or takes a nominal transaction fee from the carrier that is significantly lower than standard broker commissions.
This shift threatens to compress margins across the industry. "If a machine can analyze your risk profile better than a human and costs 99% less, the fiduciary duty to shareholders demands that companies switch," argued Sarah Chen, a senior fintech analyst at Bloomberg Intelligence. "The 13% drop we saw today in stocks like Arthur J. Gallagher isn't an overreaction; it's a repricing of future cash flows in a world where commissions vanish."
However, industry veterans are pushing back. The Council of Insurance Agents & Brokers (CIAB) released a statement emphasizing the "irreplaceable value of human relationships and strategic foresight" in risk management. They argue that while AI can process data, it cannot navigate the complex interpersonal and political nuances of large-scale corporate risk claims.
Despite the market enthusiasm for the technology—and the corresponding despair among brokerage investors—significant hurdles remain. Insurance is one of the most heavily regulated industries in the world, with a patchwork of state-level laws in the U.S. alone.
Questions regarding liability are already surfacing. If the AI agent recommends a policy that fails to cover a specific claim, who is liable? The carrier, OpenAI, or the user?
State insurance commissioners in New York and California have already signaled their intent to review the "SureAgent" application for compliance with consumer protection laws. "We will not allow innovation to bypass the safeguards that protect policyholders," noted Ricardo Lara, California's Insurance Commissioner, in a tweet shortly after the market closed.
The events of February 9, 2026, serve as a bellwether for the broader professional services economy. The "Insurance Broker Stock Selloff" is likely just the first domino. If an LLM can effectively replace a licensed insurance broker, similar disruptions may soon hit real estate agents, supply chain logistics coordinators, and travel management firms.
For Creati.ai readers, the takeaway is clear: the age of the "Agentic AI" is no longer a theoretical future. It is here, it is affecting real-world valuations, and it is reshaping the fundamental economics of service industries. As the dust settles on this trading week, all eyes will be on how the major brokerage firms pivot—whether they will fight the AI wave or attempt to acquire the very technology threatening to replace them.