When Algorithms Advise: AI’s Disruption of the Consulting Model
In October 2025, Deloitte Australia agreed to partially refund the Australian federal government after a A$439,000 report was found to contain incorrect citations and references. A corrected version disclosed that Azure OpenAI GPT-4o had been used during the drafting process, although Deloitte maintained that the report’s central findings and recommendations remained unchanged.
The controversy became a cautionary example of the consulting industry’s transition into the AI era. The issue was not simply whether artificial intelligence should be used. It was whether firms could maintain transparency, evidence standards and professional accountability while charging for work that technology could help produce much faster.
Across the consulting sector, AI is now supporting activities once carried out primarily by human teams, including data gathering, benchmarking, scenario modelling and document preparation. Major firms are integrating AI assistants and agentic workflows into their service models, promising faster delivery, greater analytical capacity and more scalable access to expertise.
But the same capabilities are also becoming available inside client organisations. Internal AI systems can increasingly gather information, compare options and produce first-stage analysis. This creates a more fundamental question for consulting firms: what can an external adviser contribute when structured analysis is no longer scarce?
The answer is likely to sit beyond the production of another report. Consulting value is moving closer to framing the right problem, challenging assumptions, interpreting organisational context and helping clients turn recommendations into implementation.
How AI Is Reshaping the Consulting Value Chain
For decades, the consulting value chain relied on the classic pyramid model: analysts gathered data, mid-level consultants synthesised findings and partners presented recommendations to executive teams. AI is now compressing parts of that structure rather than removing it entirely.
Market scans, competitor benchmarking, scenario modelling and document preparation can increasingly be completed with AI support. This reduces the time required for early-stage analysis and allows smaller teams to produce work that once depended on significant junior capacity.
The shift is not limited to how consulting firms operate. Clients are also building internal AI systems that can gather information, compare options and produce first-stage analysis. Structured analysis is becoming easier to access, which means external firms must contribute more than information production alone.
From AI Assistants to Agentic Workflows
The first wave of consulting AI tools supported individual tasks such as summarising documents, drafting reports and preparing presentations. Agentic systems go further by coordinating multiple steps, connecting different tools and acting within defined workflows.
Adoption is growing, but it remains uneven. McKinsey’s 2025 global survey found that 23% of respondents were scaling an agentic AI system somewhere in their organisation, while another 39% were experimenting. In any individual business function, no more than 10% reported scaled agent adoption.
Consulting firms are nevertheless beginning to integrate these systems into everyday delivery. McKinsey reports that active users of its internal Lilli platform have saved between two and three million hours, while engagement teams save four to six hours each week on research and presentation work.
PwC’s agent OS, launched in March 2025, connects AI agents, enterprise systems and human teams within governed workflows. Rather than replacing professional oversight, the platform is designed to combine orchestration with risk management, monitoring and human control.
These developments shift the consultant’s role towards problem framing, evidence validation, contextual interpretation and implementation. Producing the first answer is becoming faster. Determining whether it is the right answer remains a more complex responsibility.
Process Mining and AI in Operations
Beyond document production, AI is becoming part of process optimisation and operational analytics.
A 2025 insurance-industry study combined object-centric process mining with AI to automate claim-part identification across tens of thousands of records. Using event logs and relational data, the system detected anomalies and supported workflow optimisation without depending entirely on manually defined rules.
For consulting firms, the significance extends beyond the individual case. Similar approaches can support supply-chain audits, procurement analysis and healthcare service mapping by connecting recommendations to operational evidence.
This reduces dependence on manual analysis, but increases the importance of consultants who can define the right problem, test the reliability of the system and translate its findings into practical organisational change.
Real-World Government Examples
AI’s influence on consulting is not limited to private-sector transformation. Governments are also building internal analytical capabilities for work that has traditionally involved external advisory support, signalling a shift in how public policy, analysis and stakeholder engagement are delivered.
The UK Government’s “Consult” Platform
The UK government introduced Consult in 2025 as an AI tool for analysing responses to public consultations. Since then, it has moved beyond an initial pilot and into practical use across government.
In one deployment, Consult analysed more than 50,000 responses to the Independent Water Commission review in around two hours. Experts then spent 22 hours checking the results. In July 2026, the platform was also used to analyse a national consultation on children and the online world, presenting policymakers with interactive dashboards while officials reviewed and refined the themes it identified.
Traditionally, large-scale consultations relied heavily on external teams for data synthesis, sentiment mapping and policy summarisation. Consult can now complete parts of this work internally and at greater speed.
The platform does not remove the need for human review. It changes where that work happens and reduces the need to outsource every stage of the analytical process.
A Broader Policy Shift: Governments Build Internal AI Capacity
This shift forms part of a broader pattern across digitally advanced governments.
Canada launched the G7 AI Network and the GovAI Grand Challenge to develop practical, scalable AI applications for public services. The initiative brought together participants from G7 and European Union countries to work on real public-sector challenges and create solutions that could be shared across governments.
Singapore’s GovTech similarly develops internal data and AI tools to support government agencies, strengthen cross-agency analysis and improve policy and service delivery. Its tools include SENSE, an AI data assistant designed to help government officers analyse datasets without specialist technical skills.
For the consulting sector, the implication is not that governments will stop working with external firms. The relationship is becoming more hybrid, with consultancies increasingly expected to work alongside client-owned systems and internal analytical teams.
Firms that can provide governance expertise, independent challenge, implementation support and algorithmic transparency will remain valuable as public administrations expand their own AI capabilities.
Frequently Asked Questions
Will AI Replace Management Consultants?
AI is more likely to change the structure of consulting work than remove the need for consultants entirely. It can accelerate research, analysis and document production, while human judgement remains important for problem framing, interpretation, accountability and implementation.
How Is AI Changing the Traditional Consulting Model?
AI reduces the time and junior capacity required for tasks such as benchmarking, data gathering, synthesis and presentation development. This can lead to smaller teams, flatter delivery structures and greater pressure on firms to demonstrate value beyond producing analysis.
What Happens When Clients Build Their Own AI Advisory Capabilities?
Clients may be able to complete more initial research and first-stage analysis internally. External advisers will therefore need to contribute independent challenge, specialist context, governance expertise and support in turning recommendations into action.
What Are the Main Risks of Using AI in Consulting?
The main risks include inaccurate evidence, fabricated citations, weak transparency, poor data governance and unclear responsibility for AI-supported outputs. Human review must remain part of the process, particularly when recommendations affect strategic or public-sector decisions.
Where Will Consulting Value Sit as AI Adoption Grows?
Consulting value is likely to move closer to decision quality and implementation. Firms will need to show that they can frame the right problem, test assumptions, adapt recommendations to organisational realities and help clients achieve measurable outcomes.
How Metheus Can Help
We work with growth-stage and scaling businesses to develop the strategies, systems and market knowledge needed to expand with confidence. Our approach is data-oriented throughout: from feasibility assessment and market entry planning to go-to-market strategy, demand generation and strategic partnerships. We bring deep expertise in market growth and market expansion across the geographies our clients are entering, with one objective in mind: creating the conditions for sustainable, repeatable growth at scale.