FTC Sets Aside Rytr Order: What America's AI Policy Shift Means for Professional Services Globally

Last week delivered a cluster of AI governance developments on both sides of the Atlantic that professional services firms cannot afford to miss. The United States Federal Trade Commission set aside its enforcement order against AI writing tool Rytr LLC — a signal that federal AI policy is moving in a new direction. Meanwhile, California moved independently to tighten oversight, Canada committed serious public money to trustworthy AI infrastructure, and an international consortium took shape in Montreal. Taken together, these developments paint a picture of AI regulation that is fragmenting by jurisdiction while accelerating in pace. If your firm uses AI tools — and at this point, most do — you need to understand what is changing and why it matters.

What Happened With the FTC and Rytr

On 17 September 2026, the FTC set aside its order against Rytr LLC, an AI-powered writing assistant. The Commission's decision reflects a broader shift in federal AI enforcement philosophy: the current administration is deliberately pulling back from enforcement theories that risk burdening AI innovation. Commentary from Holland & Knight noted this explicitly, framing the move as part of a deliberate recalibration of how the FTC approaches AI companies.

For professional services firms, the immediate read might be: less enforcement, less risk. That is too simplistic. The FTC has not abandoned consumer protection — it has reprioritised. The same week, the comment period closed on the FTC's proposed Enforcement Policy Statement on Personalised Pricing. Announced in August 2026, this proposal makes clear that the Commission intends to pursue AI-driven pricing practices that are deceptive, unfair, or insufficiently disclosed to consumers. Firms using AI to set prices for clients — or vendors selling them tools that do — remain firmly in scope.

The practical takeaway: the FTC is becoming more selective about where it applies pressure, not more lenient across the board. Firms should not interpret the Rytr outcome as a green light for unrestricted AI deployment.

California Moves in a Different Direction

While federal enforcement softened, California's Governor Newsom signed an executive order on 18 September 2026 aimed at accelerating independent oversight of AI and advancing the development of an AI kill switch — the capacity to shut down AI systems that pose unacceptable risk.

This matters globally for a straightforward reason: California is the world's fifth-largest economy and home to most of the major AI vendors supplying tools to professional services firms. When California regulates, the products your firm relies on are affected, regardless of where your firm is headquartered. The kill switch concept in particular signals that California is serious about safety obligations, not just transparency or fairness principles. Firms in regulated industries — law, accountancy, financial services — should be watching this closely, because liability frameworks that follow from such provisions could reach into professional indemnity territory.

Canada Invests in Trustworthy AI

At the ALL IN 2026 conference in Montreal, Canada made two significant commitments that deserve attention beyond their domestic audience.

First, Canada's Minister of Artificial Intelligence and Digital Innovation announced a federal task force to guide the government's next steps on AI policy. This follows Canada's long-running effort to legislate AI through the Artificial Intelligence and Data Act, and signals that the government intends to move from deliberation toward concrete governance structures.

Second, Canada and Germany jointly committed up to $300 million in funding for LawZero, a Montreal-based non-profit focused on developing trustworthy, transparent, and safe advanced AI. Canada is contributing $150 million, Germany 100 million euros. This is not simply a research investment — it is a statement about the kind of AI infrastructure both governments want to see underpin their economies. The joint nature of the commitment also signals transatlantic alignment on AI values, even as US federal policy moves in a different direction.

On 17 September, Canada's Minister Solomon joined Mila, Mozilla, and Hypertec in announcing plans to establish a Canadian-led open-source AI consortium. Open-source AI raises its own compliance questions — about model provenance, audit trails, and accountability — that professional services firms should be factoring into their vendor due diligence.

What the Divergence Means for Professional Services Firms

The developments across North America this week illustrate a pattern that is now well-established: AI regulation is not converging into a single global framework. It is diverging, by jurisdiction and by sector. Federal US policy is loosening in some areas while tightening in others. California is pursuing its own agenda. Canada is building governance infrastructure. The EU AI Act is in implementation. The UK is operating under a principles-based approach with sector-specific guidance. The Middle East and Asia-Pacific are moving at varying speeds with differing priorities.

For an accountancy practice, a law firm, an HR consultancy, or a marketing agency operating across these jurisdictions, this means one thing above all else: compliance cannot be addressed once and filed away. It requires active monitoring, jurisdiction-by-jurisdiction mapping, and regular reassessment as the regulatory landscape shifts.

Specific pressure points for professional services firms right now include:

Personalised pricing and fee transparency. If your firm uses AI to generate quotes, set retainer fees, or customise pricing by client segment, the FTC's proposed enforcement policy on personalised pricing is directly relevant. You need to be able to demonstrate that your pricing practices are disclosed, consistent, and defensible.

AI tool procurement. The Canadian open-source AI consortium and the LawZero initiative signal growing emphasis on trustworthy, auditable AI. When you procure AI tools — for document review, HR screening, marketing automation, or financial analysis — your due diligence should include provenance, auditability, and the vendor's own compliance posture.

Operational kill switch readiness. California's executive order may seem remote, but the underlying question it raises is one every firm should be able to answer: if an AI system your firm relies on causes harm, or is found to be operating outside acceptable parameters, can you shut it down quickly and demonstrate that you had appropriate oversight in place? That is increasingly a question auditors, regulators, and clients will ask.

Cross-border data and AI governance. Firms with operations or clients across the US, Canada, EU, and UK face overlapping obligations. These need to be mapped coherently, not managed in silos.

The Week's Developments in Summary

The FTC's decision on Rytr does not mark an end to AI enforcement — it marks a shift in where enforcement energy is directed. California, meanwhile, is pushing in the opposite direction on safety. Canada is investing in trustworthy AI infrastructure and building governance structures that will shape what responsible AI use looks like in a major G7 economy. For professional services firms, the obligation is clear: understand the rules in each jurisdiction where you operate, assess your AI tools and practices against those rules, and build the internal capacity to keep pace as the framework evolves.

Work With Ops Intel

Ops Intel helps professional services firms understand and meet their AI compliance obligations — across jurisdictions, across tool types, and across the full lifecycle of AI deployment. Whether you need a compliance gap assessment, support building an internal AI governance framework, or ongoing monitoring as the regulatory landscape develops, we can help.

Find out how at opsintel.io.