Behind the AI label: how to recognise (and choose) the real HR revolution
May 28, 2026
Organisational Development in the Age of AI | Report
June 4, 2026
Organisational Development in the Age of AI | Report
Artificial intelligence is one of the main drivers of transformation for organisations today, but its real impact is still far from being fully understood, especially in the HR sphere. The public and managerial debate is dominated by enthusiasm, simplifications and often misleading narratives, leading companies to embark on superficial and ineffective adoption paths. In this context, the need for a more structured, critical and informed approach emerges strongly.
During the last Workshop Alveria Talks, “Organisational Development in the Age of AI: Designing Adaptive Systems with the AI-OD Framework”.”, held last 28 May, Filippo Cannavò offered a lsystemic approach to the role of artificial intelligence in organisational development, questioning some of the main myths surrounding its dissemination and clarifying the conditions necessary for it to generate real value.
The starting point is a an observation as simple as it is often ignored: AI is not a universal solution nor a technology that can be grafted into existing processes without changing them. On the contrary, requires a profound rethinking of organisational models, decision-making processes and internal competencies.
Through the analysis of emerging dynamics and concrete experiences observed in the market, Filippo Cannavò introduced the AI-OD (AI Organisational Development) framework, an operational model to drive the integration of artificial intelligence within organisations in a coherent, sustainable and value-oriented manner.
The context: between technological hype and organisational misalignment
In recent years, artificial intelligence has become a pervasive presence in management language and business strategies. However, this diffusion has not been accompanied by an equally deep understanding of its real implications. In fact, there is a widespread phenomenon of “background noise”, in which AI is evoked as a solution to a multiplicity of problems without a clear definition of its actual role. Within organisations, and particularly in HR functions, this results in a often superficial use of technology. The adoption of tools such as chatbots or conversational assistants is perceived as a significant advancement, when in fact it represents only a limited manifestation of the potential of artificial intelligence. This dynamic generates a deep misalignment between what the market proposes and what companies would really need. Vendor solutions tend to focus on what is easily communicated and sold, rather than on what produces structural value. As a result, many organisations find themselves implementing tools that marginally improve the user experience, but do not affect core processes or decision-making capabilities. This gap between supply and real need is one of the main obstacles to transformation.
Complicated and complex: the true application criterion
One of the most useful interpretative keys to guide the use of artificial intelligence is the distinction between complicated and complex problems. The former are characterised by a definite logic and a deterministic solution, even though they may require complex calculations. The latter, on the other hand, are inherently ambiguous, not completely predictable and influenced by multiple interdependent variables. In the case of complicated problems, the use of AI is not only superfluous, but can be counterproductive. A traditional algorithm, properly designed, guarantees greater reliability and consistency. In complex problems, on the contrary, artificial intelligence can make a significant contribution, as it is able to detect hidden patterns and support the reduction of decision complexity. This principle is particularly relevant in the HR environment, where both types of problems coexist. The ability to distinguish between these two levels is a critical competence to avoid implementation errors and maximise the value of AI.
Enabling conditions: data, infrastructure and skills
The effectiveness of artificial intelligence depends crucially on the context in which it is embedded. Without certain conditions, even the most advanced technologies are ineffective. The first essential element is the quality of the data. AI feeds on data and its ability to generate insights is directly dependent on their accuracy, completeness and updating. Added to this is the need for an adequate technological infrastructure, capable of supporting integration between systems and guaranteeing access to data in a consistent and secure manner. Without a solid architectural basis, artificial intelligence remains confined to a marginal role, limited to superficial interactions with external systems. Finally, the human factor is a decisive variable. AI does not replace competence, but amplifies it. In the absence of critical and interpretative capacities, the risk is that the generated outputs are accepted passively, without a real evaluation of their value. In this sense, artificial intelligence acts as a multiplier: it can amplify value, but also noise.
The risk of AI washing
An increasingly common phenomenon in the market is the’AI washing, the tendency to label as “intelligent” solutions that in reality are not. This often occurs through the superficial integration of external generative models, without any real adaptation work to the business context. The problem is not only semantic, but substantial. Solutions of this kind do not utilise the organisation's data, are unable to generate specific insights and do not contribute to process transformation. Their value is limited to an ancillary dimension, often already freely available through public tools. The ability to distinguish between truly intelligent solutions and simple technological rebranding therefore becomes crucial. This requires a critical approach to vendor evaluation and increased internal awareness.
From technology adoption to organisational transformation
One of the most common mistakes in the introduction of artificial intelligence is to limited to technology adoption, without intervening in processes. This approach leads to marginal results, as it inserts advanced technology within obsolete organisational models. The real opportunity offered by AI lies in the possibility of radically rethinking processes, overcoming existing inefficiencies and creating new ways of working. This implies a shift from a logic of local optimisation to a systemic vision, in which the different elements of the organisation are integrated into a coherent ecosystem.
Overcoming silos and the emergence of adaptive ecosystems
HR functions are traditionally organised in vertical silos, each focused on a specific area of activity. This structure limits the ability to generate value, as it prevents the circulation of information and the construction of an integrated vision. Artificial intelligence offers the possibility to overcome this fragmentation, creating connections between different areas and enabling a more fluid and dynamic management of processes. In this scenario, information does not remain confined within individual functions, but becomes part of an interconnected system that can continuously adapt to the needs of the organisation.
The AI-OD framework
The AI-OD framework represents a methodological approach to the design of adaptive organisational systems. It is based on a sequence of activities starting with the understanding of existing processes and ending with the targeted implementation of artificial intelligence. The first step consists of detailed process mapping, with the aim of identifying flows, interdependencies and criticalities. Subsequently, we proceed to the’analysis of inefficiencies, identifying the areas where the intervention can generate the greatest impact. It is only at this point that artificial intelligence is introduced, selectively and consistently with the available data. This approach avoids the risk of premature implementation and allows the construction of truly effective solutions, oriented towards continuous improvement.
Upcoming Alveria Talks
he next webinar, scheduled for 15 June, will be dedicated to the theme of using artificial intelligence in selection processes.
The meeting, entitled “AI in recruiting: faster or fairer decisions? The challenge for HR”will explore the role of AI in supporting decision-making, analysing its opportunities, limitations and implications for the quality of choices, between operational efficiency and human responsibility.
the contents of the webinar? Contact our team of experts or fill in the form and receive it immediately:
The full presentation of the event with additional notes from our experts
CONSULTANCY, TRAINING, HR DIGITALIZATION AND CORPORATE SOLUTIONS, DISCOVER THE ALVERIA METHOD. GET READY FOR CHANGE.
Artificial intelligence is one of the main drivers of transformation for organisations today, but its real impact is still far from being fully understood, especially in the HR sphere. The public and managerial debate is dominated by enthusiasm, simplifications and often misleading narratives, leading companies to embark on superficial and ineffective adoption paths. In this context, the need for a more structured, critical and informed approach emerges strongly.
During the last Workshop Alveria Talks, “Organisational Development in the Age of AI: Designing Adaptive Systems with the AI-OD Framework”.”, held last 28 May, Filippo Cannavò offered a lsystemic approach to the role of artificial intelligence in organisational development, questioning some of the main myths surrounding its dissemination and clarifying the conditions necessary for it to generate real value.
The starting point is a an observation as simple as it is often ignored: AI is not a universal solution nor a technology that can be grafted into existing processes without changing them. On the contrary, requires a profound rethinking of organisational models, decision-making processes and internal competencies.
Through the analysis of emerging dynamics and concrete experiences observed in the market, Filippo Cannavò introduced the AI-OD (AI Organisational Development) framework, an operational model to drive the integration of artificial intelligence within organisations in a coherent, sustainable and value-oriented manner.
The context: between technological hype and organisational misalignment
In recent years, artificial intelligence has become a pervasive presence in management language and business strategies. However, this diffusion has not been accompanied by an equally deep understanding of its real implications. In fact, there is a widespread phenomenon of “background noise”, in which AI is evoked as a solution to a multiplicity of problems without a clear definition of its actual role. Within organisations, and particularly in HR functions, this results in a often superficial use of technology. The adoption of tools such as chatbots or conversational assistants is perceived as a significant advancement, when in fact it represents only a limited manifestation of the potential of artificial intelligence. This dynamic generates a deep misalignment between what the market proposes and what companies would really need. Vendor solutions tend to focus on what is easily communicated and sold, rather than on what produces structural value. As a result, many organisations find themselves implementing tools that marginally improve the user experience, but do not affect core processes or decision-making capabilities. This gap between supply and real need is one of the main obstacles to transformation.
Complicated and complex: the true application criterion
One of the most useful interpretative keys to guide the use of artificial intelligence is the distinction between complicated and complex problems. The former are characterised by a definite logic and a deterministic solution, even though they may require complex calculations. The latter, on the other hand, are inherently ambiguous, not completely predictable and influenced by multiple interdependent variables. In the case of complicated problems, the use of AI is not only superfluous, but can be counterproductive. A traditional algorithm, properly designed, guarantees greater reliability and consistency. In complex problems, on the contrary, artificial intelligence can make a significant contribution, as it is able to detect hidden patterns and support the reduction of decision complexity. This principle is particularly relevant in the HR environment, where both types of problems coexist. The ability to distinguish between these two levels is a critical competence to avoid implementation errors and maximise the value of AI.
Enabling conditions: data, infrastructure and skills
The effectiveness of artificial intelligence depends crucially on the context in which it is embedded. Without certain conditions, even the most advanced technologies are ineffective. The first essential element is the quality of the data. AI feeds on data and its ability to generate insights is directly dependent on their accuracy, completeness and updating. Added to this is the need for an adequate technological infrastructure, capable of supporting integration between systems and guaranteeing access to data in a consistent and secure manner. Without a solid architectural basis, artificial intelligence remains confined to a marginal role, limited to superficial interactions with external systems. Finally, the human factor is a decisive variable. AI does not replace competence, but amplifies it. In the absence of critical and interpretative capacities, the risk is that the generated outputs are accepted passively, without a real evaluation of their value. In this sense, artificial intelligence acts as a multiplier: it can amplify value, but also noise.
The risk of AI washing
An increasingly common phenomenon in the market is the’AI washing, the tendency to label as “intelligent” solutions that in reality are not. This often occurs through the superficial integration of external generative models, without any real adaptation work to the business context. The problem is not only semantic, but substantial. Solutions of this kind do not utilise the organisation's data, are unable to generate specific insights and do not contribute to process transformation. Their value is limited to an ancillary dimension, often already freely available through public tools. The ability to distinguish between truly intelligent solutions and simple technological rebranding therefore becomes crucial. This requires a critical approach to vendor evaluation and increased internal awareness.
From technology adoption to organisational transformation
One of the most common mistakes in the introduction of artificial intelligence is to limited to technology adoption, without intervening in processes. This approach leads to marginal results, as it inserts advanced technology within obsolete organisational models. The real opportunity offered by AI lies in the possibility of radically rethinking processes, overcoming existing inefficiencies and creating new ways of working. This implies a shift from a logic of local optimisation to a systemic vision, in which the different elements of the organisation are integrated into a coherent ecosystem.
Overcoming silos and the emergence of adaptive ecosystems
HR functions are traditionally organised in vertical silos, each focused on a specific area of activity. This structure limits the ability to generate value, as it prevents the circulation of information and the construction of an integrated vision. Artificial intelligence offers the possibility to overcome this fragmentation, creating connections between different areas and enabling a more fluid and dynamic management of processes. In this scenario, information does not remain confined within individual functions, but becomes part of an interconnected system that can continuously adapt to the needs of the organisation.
The AI-OD framework
The AI-OD framework represents a methodological approach to the design of adaptive organisational systems. It is based on a sequence of activities starting with the understanding of existing processes and ending with the targeted implementation of artificial intelligence. The first step consists of detailed process mapping, with the aim of identifying flows, interdependencies and criticalities. Subsequently, we proceed to the’analysis of inefficiencies, identifying the areas where the intervention can generate the greatest impact. It is only at this point that artificial intelligence is introduced, selectively and consistently with the available data. This approach avoids the risk of premature implementation and allows the construction of truly effective solutions, oriented towards continuous improvement.
Upcoming Alveria Talks
he next webinar, scheduled for 15 June, will be dedicated to the theme of using artificial intelligence in selection processes.
The meeting, entitled “AI in recruiting: faster or fairer decisions? The challenge for HR”will explore the role of AI in supporting decision-making, analysing its opportunities, limitations and implications for the quality of choices, between operational efficiency and human responsibility.