The question is not whether to use AI in the company, but how. These are the proposals from Microsoft, Gartner, McKinsey, and Accenture, along with their strengths and weaknesses.
Generative AI can be much more than a tool, and the key is how to use it to organize work, make decisions, and develop talent.
There are four frameworks worth knowing for their influence, for how well they complement each other, and for the differences between them:
- Microsoft (Frontier Firm).
- Gartner (ethical, human-centered AI).
- McKinsey (skills-based augmentation).
- Accenture (Reinvention + Trust).
1) Microsoft - "Frontier Firm": humans + agents working side by side
The idea
Microsoft proposes moving from "people who use tools" to hybrid teams of people and AI agents.
It outlines a three-phase journey: first, AI as an assistant; then as a "digital colleague" that carries out complete tasks; and finally, as an agent that, under human direction, executes entire processes.
It introduces new roles, such as "agent manager" or "head of agents," and calls for reorganizing work around projects and outcomes rather than rigid hierarchies.
How it is applied
This involves redesigning processes to "decouple" people from repetitive tasks, training them to design or supervise agents, and rethinking the org chart toward dynamic "work charts."
It presents cases of impact on productivity and timelines, along with business metrics to measure improvement when AI is integrated into the actual workflow.
Strengths
Operational traction: it provides paths, roles, and KPIs, as well as ROI and productivity data that make it a highly actionable framework.
Weaknesses / risks
Risk of technological dependence and of pushing too fast without maturing governance and skills; cultural resistance if the change is perceived as dehumanizing.
2) Gartner - Ethical, human-centered AI: governance to scale with confidence
The idea
Gartner puts trust at the center. Its AI TRiSM framework (Trust, Risk and Security Management) proposes controls across the model lifecycle: governance, transparency, bias management, security, and compliance. The thesis is that without the trust of employees, clients, and regulators, adoption is not sustainable.
How it is applied
Start by defining policies, roles, and responsibilities, audits and risk metrics, and training that is not only technical but also ethical. This framework speeds up time-to-market and raises client trust if implemented well.
Strengths
It reduces reputational and legal risks and creates the conditions to scale in an orderly and lasting way.
Weaknesses / risks
It can slow things down if it becomes bureaucratic: "paralysis by analysis" and complexity in standardizing policies across different contexts. Care is also needed to avoid "ethics washing" if there is no real oversight.
3) McKinsey - Skills-based augmentation: talent as a lever
The idea
The focus is not the tool but skills. AI serves to identify, develop, and deploy competencies in real time (skills-based recruitment, internal mobility, personalized learning). The goal is a more adaptable and diverse workforce.
How it is applied
Shift HR toward an internal skills marketplace: a skills inventory, tiered career paths, and AI-assisted continuous training. It improves person-role fit and retention, opening opportunities for non-traditional profiles.
Strengths
It improves the agility and fit of talent; it promotes diversity and inclusion; and it reduces turnover costs by offering clear progression based on competencies.
Weaknesses / risks
It requires a deep cultural change (managers and processes), good assessment metrics, and avoiding bias if degree requirements are removed without replacing them with objective tests. It requires a high initial effort.
4) Accenture - Reinvention + Trust: transforming the company and the employee experience
The idea
A dual movement: "Total Reinvention" (business model, processes, and an "AI-native" culture) and building trust so that people feel "Net Better Off" (coming out ahead) with AI.
How it is applied
Orchestrated transformation: maturity diagnostics, centers of excellence, pilots with the business, large-scale training, and ambassador programs ("responsible AI champions"). Both business results and the employee experience are measured.
Strengths
A holistic vision (business + culture + people) with efficiency and growth cases, and an explicit focus on internal acceptance.
Weaknesses / risks
The complexity and magnitude of the change: it demands leadership, investment, and very good risk management; otherwise, it gets diluted or becomes too abstract.
Cross-framework comparison
- Main purpose
- Microsoft: scale capacity with agents and operational redesign.
- Gartner: ensure trust and compliance to be able to scale.
- McKinsey: compete through talent (skills) and real internal mobility.
- Accenture: reinvent the company and culture while measuring whether people are better off.
- Levers
- Microsoft: processes, agents, operational ROI.
- Gartner: governance frameworks (AI TRiSM), controls, and risk metrics.
- McKinsey: skills inventory, career paths, training.
- Accenture: centers of excellence, pilots with the business, and AI ambassadors.
- Role of people
- Microsoft: design and supervise agents (e.g., head of agents).
- Gartner: co-creators of policies and guardians of ethics.
- McKinsey: protagonists of augmentation: growing through competencies.
- Accenture: copilots of change; the aim is for people to "come out ahead."
- Typical risks
- Microsoft: technological dependence and cultural shock if things move too fast.
- Gartner: slowness/bureaucracy or "window-dressing ethics" if there is no real execution.
- McKinsey: measuring skills well and sustaining the effort (not stopping at the initial momentum).
- Accenture: complexity and the risk of dispersion if the "reinvention" does not take root.
- Which KPIs they track
- Microsoft: productivity, throughput, process ROI.
- Gartner: compliance, incident reduction, and client/employee trust.
- McKinsey: time to fill vacancies, internal mobility, retention, D&I.
- Accenture: growth + experience metrics ("Net Better Off").
Comparison table
| Criterion | Microsoft (Frontier Firm) | Gartner (ethical, human-centered AI) | McKinsey (skills-based augmentation) | Accenture (Reinvention and trust) |
|---|---|---|---|---|
| Purpose | Scale with agents and operational redesign | Scale with trust and governance | Compete with talent and skills | Transform business and culture, "Net Better Off" |
| Levers | Processes, agents, ROI | AI TRiSM, policies, audits | Skills inventory, career paths, training | CoE, pilots, ambassadors, maturity |
| Role of people | Design and supervise agents | Co-design rules and controls | Protagonists of growth | Copilots of reinvention |
| Strengths | Operational roadmap and KPIs | Reduces risks and supports sustainable adoption | Talent agility, D&I | Holistic business-culture-people vision |
| Risks | Tech dependence, cultural shock | Slowness/bureaucracy, "ethics washing" | Measuring and scaling cultural change | Complexity/abstraction, dispersion |
| Typical KPIs | Productivity/process ROI | Trust, compliance, incidents | Retention, mobility, time-to-fill | Growth + employee experience |
Which one should I choose?
The best approach is probably not to look for the silver bullet, but to combine them sensibly:
- Quick value delivery using Microsoft's operational logic (pilots with agents in processes that hurt the most), because it sets expectations, roles, and metrics from day one.
- Avoiding surprises by applying Gartner's safety net: clear policies, lightweight TRiSM to get started, and simple risk metrics (bias, privacy, acceptable use).
- Investing in skills with McKinsey's perspective: skills inventory, career paths, and AI feedback for learning in the flow of work.
- Developing culture and trust in Accenture's style: communicating the purpose, measuring the experience ("are people better off?"), and creating internal ambassadors who accelerate adoption.
I believe the best advice is to learn the frameworks, assess your own reality, and based on that, decide with your own judgment which piece to apply first. Start small, measure well, communicate better, and scale wisely. Learn quickly without burning through trust or talent.
Learn more
Microsoft - "Frontier Firm"
Gartner - AI TRiSM (ethical AI, risk and security)
- Official glossary: AI TRiSM
- Article: AI Trust and AI Risk
- Article: AI Ethics: Enable AI Innovation With Governance Platforms
Comments (1)
Log in to your Scrum Manager area to comment
Jorge Sánchez López ★★★
17 Oct 2025 12:45Thank you for sharing this so thorough analysis, Juan. As a professional dedicated to organizational transformation from agile and humanistic approaches, I would like to offer a critical and down-to-earth perspective on these four frameworks.
The first thing I appreciate is that AI is not treated as a fad, but as a structural change that requires reflection, design, and conscious leadership. In that sense, I fully agree with the idea that there is no "silver bullet" and that the most sensible approach is to combine approaches according to the reality of each organization.
It is not about choosing a framework, but about combining them with good judgement.
- Microsoft offers a clear, operational roadmap, but be careful not to run ahead of the culture. AI must not dehumanize work.
- Gartner puts the focus on trust. Fundamental. But ethics is not imposed through policies; it is cultivated through practice. Are we ready to govern with awareness?
- McKinsey champions talent and skills. Very aligned with agile thinking. But measuring competencies well and avoiding bias is key so we do not unintentionally exclude people.
- Accenture proposes a total reinvention. Ambitious and necessary, but it demands mature leadership and flawless change management. Do we have the internal ambassadors to make it possible?
As in any transformation, success lies not in the framework, but in how it is applied, who leads it, and how it is communicated.
From my experience, I recommend starting by:
- Listening to people before redesigning processes.
- Training in critical thinking, not just in tools.
- Measuring the human experience, not just the ROI.
- Iterating with humility, learning from what does not work.
Thank you for opening this space for reflection. AI is not just a technology; it is an opportunity to reimagine work with more humanity, purpose, and collaboration.
Gracias por compartir este análisis tan completo, Juan. Como profesional dedicado a la transformación organizacional desde enfoques ágiles y humanistas, me gustaría aportar una mirada crítica y cercana sobre estos cuatro marcos.
Lo primero que celebro es que no se plantea la IA como una moda, sino como un cambio estructural que requiere reflexión, diseño y liderazgo consciente. En ese sentido, coincido plenamente con la idea de que no hay una “bala de plata” y que lo más sensato es combinar enfoques según la realidad de cada organización.
No se trata de elegir un marco, sino de combinarlos con criterio.
- Microsoft nos ofrece una hoja de ruta operativa clara, pero cuidado con correr más que la cultura. La IA no puede deshumanizar el trabajo.
- Gartner pone el foco en la confianza. Fundamental. Pero la ética no se impone por políticas, se cultiva desde la práctica. ¿Estamos preparados para gobernar con conciencia?
- McKinsey apuesta por el talento y las skills. Muy alineado con el pensamiento ágil. Pero medir bien las competencias y evitar sesgos es clave para no excluir sin querer.
- Accenture propone una reinvención total. Ambicioso y necesario, pero exige liderazgo maduro y una gestión del cambio impecable. ¿Tenemos los embajadores internos que lo hagan posible?
Como en cualquier transformación, el éxito no está en el marco, sino en cómo se aplica, quién lo lidera y cómo se comunica.
Desde mi experiencia, recomiendo empezar por:
- Escuchar a las personas antes de rediseñar procesos.
- Formar en pensamiento crítico, no solo en herramientas.
- Medir la experiencia humana, no solo el ROI.
- Iterar con humildad, aprendiendo de lo que no funciona.
Gracias por abrir este espacio de reflexión. La IA no es solo una tecnología, es una oportunidad para reimaginar el trabajo con más humanidad, propósito y colaboración.
Automatic translation · View original