Salesforce, BCG, and the Centre for Public Impact provide practical actions to cut through the complexity of AI approval processes and unlock new public value.
Australian public sector agencies are being urged to move beyond just the high level principles in their risk and assurance frameworks guiding AI implementation, simplifying some of the existing complex and duplicative approval processes.
A new global report released today by Boston Consulting Group (BCG), Salesforce, and the Centre for Public Impact highlights a major opportunity to accelerate government AI delivery, while still meeting citizens’ high expectations for trust and safety.
While many governments both locally and overseas have established strong foundations with robust, high-level AI ethics principles, the next phase of maturity lies in streamlining day-to-day operational workflows.
The report, Trust Imperative 5.0: Building Trust in Government through Practical AI Assurance, suggests governments ensure AI assurance frameworks do not slow down progress unnecessarily because they need to keep up with societal expectations and adoption, or risk eroding public confidence and trust.
To earn genuine public trust in AI, we need to build for the citizen outcome which includes keeping data safe, producing accurate results, and delivering on the deeply human side of government service
Gisele Kapterian, Public Sector Strategy Lead ANZ, Salesforce
Governments should take a phased and lifecycle-based approach, with the report highlighting how the public sector can boost improvements to service delivery while carefully managing risk.
Addressing the operational bottlenecks comes at a critical time. The latest 2026 global data from BCG shows that weekly AI usage among citizens has risen by more than 25% since 2024. Proficiency is also increasing, creating acceptance, the social license and driving demand and expectations for public and private sector organisations to incorporate and use AI more.
“The question is no longer whether to govern AI responsibly, but whether our assurance processes are actually helping the adoption of technology that could transform public services,” said report co-author Gisele Kapterian, Public Sector Strategy Lead ANZ, Salesforce.
“At its heart, assurance is about building confidence and competence. To earn genuine public trust in AI, we need to build for the citizen outcome which includes keeping data safe, producing accurate results, and delivering on the deeply human side of government service.
“Governments that get this right won’t just be more efficient, they’ll be more trusted. And in an era where confidence in public institutions is hard-won, that matters more than ever,” she concluded.
Smarter Controls for the Agentic AI Era
The report introduces a defined operating model that separates organisational roles and clear accountabilities for enforcing engineering guardrails, ensuring that public sector organisations can break the gridlock faster and see real progress sooner.
“Applying the principles and frameworks in practise has proven more challenging than necessary for real-world practitioners in the public service” added Miguel Carrasco, Managing Director and Senior Partner at Boston Consulting Group and Chair at the Centre for Public Impact ANZ. “AI projects are delayed or stalled, because frameworks are out-of-date, unclear, duplicative or the process is too hard to navigate.”
“AI could generate significant public value, but without a simpler, more practical approach, many of those benefits risk being left on the table.”
The report provides nine practical suggestions to make AI assurance more effective and responsive, moving at the speed of the technology itself and preventing valuable AI tools from remaining permanently trapped in the pilot stage. These include:
- Smarter Triage and Setup: Simplify risk triage and clarify roles for clearer accountability to fast-track low-risk productivity tools within established guardrails and updated AI frameworks.
- Streamlined Delivery Integration: Adopt a staged lifecycle approach, streamline review workflows and reduce duplication with reusable artifacts.
- Capability Uplift and Continuous Evaluation: Build baseline AI literacy across all staff and shift away from passive compliance forms to active lifecycle monitoring.
Moving from Frameworks to Practice
The report highlights lessons learned and examples from around the world. For instance, New South Wales — which the report singles out for leading the nation in governance maturity — successfully compressed a specialist-heavy review process that took roughly 40 hours down to an intuitive, 15-minute inherent-risk triage.
Internationally, there are clear examples of governments turning abstract principles into practical transparency that eliminates duplicative department tasks. For instance, Japan has replaced rigid preapproval mandates with a flexible, tiered report-and-review pathway, while the United Kingdom uses a mandatory central transparency standard to clear a predictable path for AI usage across public departments.
“The aim is to make AI governance better and more efficient, so that more AI projects make it from proof-of-concept, to pilot, to production more quickly,” concluded Miguel Carrasco. “If we do that, then we will all benefit from the enormous innovation, productivity and value that AI – done well and responsibly – can bring to policy, regulation and service delivery.”
More Information:
- Access the full Trust Imperative 5.0 report
- Find out more about the 2026 Public Sector Summit
- Learn about the rise of the Agentic Government
- Catch up on the launch of Agentforce for Public Sector
