For the past 30 years, we have have been accustomed to transacting online, from retail purchases to shopping for insurance, cars, and services. Since the end of last year, though, it is not longer just humans that perform these transactions. AI agents have entered the business landscape through agentic browsers and plugins such as OpenAI’s Atlas [already sunset], Perplexity Comet, or Claude in Chrome. These agents are making it easier for AI assistants to navigate web sites that are optimized for humans, and they are increasing the level of autonomy and automation to accomplish tasks and transactions.
Agentic commerce is the next wave of innvoation built on this technology, and it is arriving faster than the trust infrastructure it depends on. That is why the standard consent model requires an update as well. IEEE 7012, a new standard is proposing to change that. What terms and privacy terms do you agree to, and what does your agent need to abide by when transacting on your behalf? Here’s what we talked about…
Agentic Commerce is Outrunning Consent
OpenAI has partnered with PayPal and Walmart. Starbucks now markets ordering your latte through ChatGPT. OpenAI has floated learning about your banking and financial assets so it can advise you. None of these are edge cases anymore. They are the early shape of a world where an AI agent, yours or a vendor’s, is doing the transacting, and each one on its own looks like a convenience.
Add them up and you get something else: a default assumption that agents can act on our behalf before anyone has agreed on what that agent is actually allowed to promise. If your agent goes out and books you a flight that costs twice what you would have paid yourself, try disputing that after the fact. The agent acted “on your behalf,” and the terms of service you clicked through years ago never anticipated that scenario.
This is the gap leaders need to see now, not after a trust incident makes it visible. If your organization is deploying agents to negotiate, procure, or transact with customers or suppliers, the terms governing what those agents can commit to on your behalf deserve the same scrutiny you would give a new employee’s signing authority. Although technology is moving, the governance around it is not moving at the same pace.
Get your copy: The HUMAN Agentic AI Edge
Organizations are racing to deploy Agentic AI, yet few are ready for the risks that emerge when employees use AI without structure, standards, or oversight.
The HUMAN Agentic AI Edge offers leaders a practical blueprint for building accountable AI-ready teams that consistently produce high-quality results. Drawing on real-world knowledge and insights from interviews with more than 50 AI leaders and experts, Andreas Welsch shows how to combine human judgment with Agentic AI capabilities to achieve the performance many organizations expect but rarely deliver. This book prepares you to shape the next generation of AI-ready teams delivering high-quality results with high accountability.
A Real Standard for Machine-Readable Consent Already Exists
There is already a working answer, and it has a name. IEEE 7012, nicknamed “MyTerms,” went live earlier this year. It grew out of Project VRM, a research effort based at Harvard’s Berkman Klein Center going back to 2007, and it now sits under a nonprofit called Customer Commons, modeled directly on how Creative Commons standardized licensing for creative work.
The idea is straightforward. Instead of one all-or-nothing terms of service that you accept before you even see what a company offers, MyTerms translates legal terms into a form both people and machines can read and act on. There are already five preset templates, ranging from a bare “service delivery only, no data captured” tier, similar to walking into a store and paying cash, up to fuller arrangements where a customer chooses to share more in exchange for more personalized service. Instead of a website tracking you indefinitely the moment you land on it, the agreement becomes explicit and, in principle, negotiable between your agent and the vendor’s.
The interaction protocols that let two agents actually negotiate these terms are still being built, and Europe and the US are approaching it from opposite directions, one regulation-first, one market-first. But “MyTerms” is a real, IEEE-backed standard you can read about today at myterms.info, and it is worth understanding before it shows up in a vendor contract or before your own agents need to honor one.
Consent is a Business Advantage
CX and commerce leaders might wonder why a vendor would ever agree to it. The answer is clear. A customer operating under a clear, mutual agreement shares more real, useful context than a customer who feels surveilled. For example, a customer will not hand their retirement savings strategy to just anyone claiming to be a financial expert. They want to work with a certified planner, and once that bar is met, they will share the specifics freely.
Carrying that same framing into your own planning. Better customer data that improves service and personalization tends to come from customers who trust the terms of the exchange, not from customers whose behavior is passively harvested without their attention. If your agentic commerce strategy is built on collecting more data by default, you may be optimizing for volume over the quality of context that actually drives better outcomes. Treating consent as a lever for better data, rather than only a compliance obligation, changes how you’d design the agent-to-agent handshake from the start.
Summary
Agentic commerce is moving ahead of the consent infrastructure meant to support it, visible in how quickly agents are being wired into payments, retail, and even financial advice. IEEE 7012, or “MyTerms,” is a real, standard that turns legal terms of service into something both people and their AI agents can read, negotiate, and honor, with tiered presets instead of all-or-nothing tracking.
If your organization is putting agents in front of customers, or letting agents act on your customers’ behalf, this is the moment to ask who owns the terms of that exchange, and whether your current terms of service would survive being read by a machine. What would your customers be willing to share with your agents if they trusted the terms as much as they trust walking into a store and paying cash?
Equip your team with the knowledge and skills to leverage Agenti AI effectively. Book a consultation or workshop to accelerate your company’s AI adoption.
Join my bi-weekly live stream and podcast for leaders and hands-on practitioners. Each episode features a different guest who shares their AI journey and actionable insights. Learn from your peers how you can lead artificial intelligence, generative AI, agentic AI, and automation in business with confidence.
Not sure if we will ever get there with agentic purchases. Having AI suggest vs fully handing over that trust is a huge jump. 74% of consumers trust personal AI agents for general shopping tasks. But just 9% trust AI to autonomously complete payment transactions. (Accenture, 2026).
Not sure if we will ever get there with agentic purchases. Having AI suggest vs fully handing over that trust is a huge jump. 74% of consumers trust personal AI agents for general shopping tasks. But just 9% trust AI to autonomously complete payment transactions. (Accenture, 2026).