As one of its first acts, the Trump administration in January signed Executive Order 14179, removing previous regulatory guardrails for artificial intelligence and placing the responsibility for this transformative technology squarely with the private sector.
The private sector needs to embrace this duty, because the rapid development of so-called “AI agents,” which promise to transform the way consumers travel, shop and even receive medical care, is going to demand cooperative industry action to ensure open and competitive markets.
History has shown us that network effects — where the value of a platform grows as more people use it — can lead to significant market concentration. This dynamic can enable a small number of players to establish dominance and lead to high barriers to entry for newcomers.
Artificial intelligence is fated to follow this same trend. In fact, we are at a critical juncture where the same dynamics that created digital monopolies in the past are beginning to take hold — only faster, and with broader implications.
The next frontier of AI is not just what we see today: large language models or image generators. It is autonomous agents: AI systems acting on our behalf in everyday transactions. These agents will manage our schedules, compare insurance plans, negotiate purchases and more. They promise to make our lives easier by operating behind the scenes to save us time and money, and spare us cognitive load.
And their adoption is happening quickly. A recent survey by Cloudera found that 96 percent of IT leaders plan to increase their use of AI agents in the coming year, with nearly half already seeing them as a key competitive advantage. Moreover, Walmart’s announcement that it plans to start interfacing with AI shopping agents signals that we are already at the beginning of a structural shift in how decisions are made online.
But the benefits of “agentified” commerce won’t materialize in full force unless we take affirmative steps to protect the promise of AI. Without standards to ensure open participation and fair competition, the agent-driven marketplace could become yet another closed system dominated by the few companies that have the resources and infrastructure to scale quickly. Smaller businesses could find themselves locked out. And consumers could find themselves at the mercy of hidden algorithms that aren’t working in their best interests.
Imagine this near-future scenario: you ask your AI assistant to plan a weekend trip to Napa Valley. It scans dozens of options, compares prices, negotiates availability, and returns with what seems like the best result. But what if it only considers providers that have exclusive deals or undisclosed business relationships with its parent company? What if smaller, independent options never even get a chance to compete? Now imagine this across other scenarios, such as finding a new healthcare provider or renegotiating your internet plan.
History has shown that these fears are not speculative. But we’ve navigated similar challenges before. The internet as we know it runs on mostly open, decentralized standards that allow anyone to build and compete on a level playing field. In hardware, protocols such as USB, Wi-Fi and Bluetooth have enabled interoperability across brands and devices, helping innovation flourish without locking out newcomers.
We should bring this same thinking to AI, and how the dynamic ecosystem of different AI tools and systems will interact — both with consumers and with each other. The future of AI-driven commerce demands open standards that ensure not just interoperability between agents but equitable access to marketplaces, so that a startup’s offering can be just as discoverable as that of a global enterprise.
One way to operationalize this vision is through the creation of a voluntary open AI agent registry. In this system, any business, regardless of size, could register its AI agents using standardized protocols. Consumer-facing AI assistants could query this registry to identify relevant service providers, ensuring that small players are part of the ecosystem from the start — and that agents are who they say they are, not scammers.
In the case of travel, for instance, this would allow a locally owned B&B to appear in the same search as a multinational hotel chain. The user’s agent could negotiate with both, compare deals, and surface the best option — not the one with the biggest marketing budget, or one that owns the cloud platform on which it is hosted.
To make this happen, we need leadership, from both regulators and industry. Standards don’t have to come from government mandates. In fact, many of the most successful ones have emerged from coalitions of private-sector leaders, academics and technologists, such as the Financial Data Exchange, which helped define open protocols for sharing consumer financial data securely across banks and fintechs.
The benefits of getting this right are hard to overstate. Consumer trust is foundational, not just to AI adoption but to long-term confidence in digital systems that increasingly act on our behalf. For businesses, especially smaller ones, well-defined standards level the playing field and reduce integration hurdles, enabling broader participation in the digital economy. And for the broader ecosystem, it ensures a competitive, innovation-rich environment where value — not gatekeeping — wins.
The best markets are free, open and competitive, and that should be true especially when transactions are handled by AI. We don’t have to repeat the mistakes of the past. We can build an AI ecosystem that is open, competitive and fair from the start. But that requires intention, collaboration and urgency. e, and that should
The AI economy is moving fast. Let’s keep it moving in the right direction.
Benjamin Wiener is the global head of Cognizant Moment, the digital and customer experience arm of U.S. professional services firm Cognizant.
