The global skills and competency framework for the digital world
SFIA - a framework for AI skills
Challenges in defining AI skills
The rapidly evolving nature of AI technologies presents several challenges:
Pace of change: AI technologies advance quickly, making static frameworks difficult to maintain
Breadth and depth: Balancing comprehensive coverage without excessive specificity
Integration: Complementing existing professional skills rather than creating siloed specialisations
Industry variation: Different sectors adopt and apply AI in diverse ways
Ethical considerations: Addressing significant ethical and legal implications
User needs: Maintaining practicality for current users while adapting for future developments
SFIA's approach to AI skills
SFIA provides a flexible framework describing professional skills for working with AI without prescribing specific technologies:
Focuses on enduring professional capabilities adaptable to any AI platform or tool
Maintains a practical, industry-proven approach that has supported employers through previous technological transformations
Complements specialist AI frameworks, data literacy standards, and governance models (such as those published by government bodies)
Integrates AI skills within SFIA's established seven-level framework, avoiding artificial boundaries between specialisms
Benefits for organisations
This approach enables organisations to:
Develop sustainable AI capabilities that evolve with technological advancement
Support workforce mobility with clearer career pathways
Understand how AI capabilities connect with and enhance existing professional skills
For organisation's using GenAI and LLMs for people and skills management - SFIA provides structured, expert-curated content to help mitigate risks
SFIA 9 emphasises capabilities that remain relevant despite rapid technological change, introducing new skills only where specific AI expertise is required.
Using SFIA for managing AI/ML skills, task analysis and people
A skills-based strategy is crucial to get the most out of the potential of artificial intelligence (AI) and machine learning (ML) technologies.
Whether it involves deploying AI-powered tools, re-engineering business processes, building machine learning models, or scaling systems, SFIA's structured skills framework helps organisations identify the key capabilities needed
For organisations adopting AI/ML technologies, SFIA provides a clear view of required skills—supporting effective deployment, optimisation, and ongoing maintenance of intelligent systems. Its levelled definitions aid in mapping skills , ensuring the right expertise is applied at each stage.
By prioritising practical skills, organisations broaden their talent search, identifying individuals who can contribute to AI/ML projects from various backgrounds and experiences.
SFIA's extensive and human curated task-oriented descriptions and responsibility levels can also act as:
A checklist for spotting potential tasks and activities where AI can automate, assist, augment
Every SFIA skill description is already phrased in “tasks you perform”. Those verbs (e.g. collect, analyse, approve, monitor…) help teams pick out which steps could be automated, assisted or augmented by AI.
A template for documenting what stays human
The same descriptions can be reused to define the vital human tasks and responsibilities once an AI tool and a re-engineered process is in place.
A responsibility gauge
SFIA’s seven responsibility levels (described by autonomy, influence, complexity) provide a ready-made scale for asking questions such as “Are we comfortable delegating a Level 3 ‘enable’ activity to AI, or do we keep a human at Level 5 ‘ensure and advise’ in the loop
Why choose SFIA?
Openness, collaboration, interoperability and availability to all employers and professionals worldwide sets the SFIA framework apart from similar frameworks in the industry.
Unlike many other frameworks, SFIA is open and accessible to all.
For most employers and all personal users, there are no fees or subscriptions required to access the content.
The framework is easy to access in many formats - web, pdf, xlsx, JSON, RDF
Updates to the framework are made transparently, we don't do "behind closed doors" updates
Mitigating AI Risk in people and skills management
Generative-AI tools that draft job descriptions, screen candidates, or recommend learning paths fall squarely into the "high-risk" category for worker-management systems under AI governance frameworks. That means employers must show robust data quality, transparency, explainability, and bias-mitigation controls—all hard to achieve if the model is left to improvise.
SFIA's skill and level descriptions provide structured, expert-curated content to help mitigate these risks.
An illustration - of the tasks, activities and SFIA skills needed to re-engineer business processes to use AI
Finding AI - related skills in SFIA 9
Use the links below to find curated content describing the application of SFIA skills to the task, activities and responsibilities needed to work effectively with AI and Machine learning.
SFIA provides a common language throughout the skills management cycle
7 levels describing increasing responsibility, accountability and impact
SFIA's seven levels of responsibility provide a a structured framework for AI and ML skills, covering everything from foundational tasks to strategic leadership in intelligent systems.
This approach aligns job roles with real-world AI and ML demands, ensuring that practitioners at all levels possess the necessary expertise to build, operate, and enhance AI-driven technologies.
It supports the development of critical capabilities, from basic AI tool usage and data preparation to advanced model development, operational deployment, ethical considerations, and strategic decision-making.
SFIA enables focused professional growth, spanning the full spectrum of AI/ML skills—supporting career progression from hands-on data analysis and automation tasks to expert-level model optimisation, governance of intelligent systems, and leadership in AI strategy and innovation.
SFIA skills encompass the broad AI and ML landscape
Spanning the full spectrum of digital, data, technology, and ethical considerations, SFIA's skill descriptors cater to both specialised and generalist roles within the AI and ML domain.
They detail the critical skills required for roles focused on the development, deployment, and maintenance of AI and ML technologies. This extends to a variety of other technical and business disciplines, including data science, algorithm optimisation, AI ethics, architecture, security, leadership, supplier management, organisational change, and people and skills management.
This enables a unified approach to skills for AI-driven transformation, ensuring organisations can identify, develop, and manage the talent required to harness the potential of intelligent systems.
Click image to expand.
Applying SFIA's uniform skills language to AI and ML
Provide a structured, consistent language for AI/ML skills so organisations can map roles, conduct skills assessments and plan development pathways.
Bridge specialist and non-specialist communities: highlights that non-specialist professionals also require foundational AI literacy to apply the technology responsibly within their domain.
Act as a quick-reference gateway to the full SFIA framework, with url short codes that link directly to detailed descriptions online.
SFIA provides a structured and consistent approach to defining skills. Each skill is clearly described, supplemented by guidance notes, and detailed level-by-level practice descriptions that align with the framework's 7 levels of responsibility.
This uniform structure ensures ease of navigation and understanding, seamlessly integrating professional skills with behavioural factors to outline comprehensive role expectations.
The consistent detail across all levels ensures robustness, allowing for precise skills and competency assessment.
The clarity in describing the specific nature of technology and business roles at every responsibility level makes it invaluable for developing and benchmarking AI/ML and other related digital capabilities within an organisation.
Click image to expand.
How SFIA works - Levels of responsibility and skills
Explore SFIA’s seven levels of responsibility, outlining the progression of responsibility, accountability and impact, and how professional skills and generic attributes define competence and support professional growth.
The value of SFIA's business skills and behavioural factors in AI adoption
Organisations exploring AI and developing and operating AI-based solutions need more than just technical expertise—they require well-rounded professionals with strong business skills and behavioural attributes. By combining these generic attributes with professional skills, SFIA provides a complete picture of what is expected from individuals at every level. The generic attributes also help organisations assess and plan for professional growth, making it easier to map career progression and align roles with business goals.
You can think of SFIA as an open-source project for professional skills.
We welcome contributions from the wider community to ensure our framework evolves with the needs of the industry. Because SFIA is not proprietary to specific vendors or industry bodies - anyone can provide feedback, suggestions, and contributions.
Field-tested approach over 20+ years: Used by practitioners, not a theoretical model
Established update process: Regular revisions with contributors, a global design authority, beta testing, structured releases and translations. Skills for AI, data and machine learning have been introduced since SFIA version 7 in 2018.
Global adoption: Across government, corporate, and education sectors in 12 languages
Expanding resources over time: Growing guidance materials and industry partnerships
Organic growth: Primarily spread through word-of-mouth success
