Two organisations, both facing the imperative to reinvent themselves. On one side, the outsized ambition of MD Anderson Cancer Center — one of the world's most prestigious hospitals — partnering with IBM to attempt the impossible: eradicating cancer through the power of Watson artificial intelligence. This is the story of a $62 million project designed to substitute a digital oracle for human medical expertise. On the other side, Microsoft in 2014: an ageing empire, perceived as arrogant, losing ground to Google and Apple. Under Satya Nadella, the company does not bet on a new miracle product, but embarks on a deep introspection about its internal culture. This is the story of a renaissance built not on computer code, but on the rewriting of the company's social code.
This comparison illuminates how the most advanced technology, without alignment of human behaviours and organisational structures, leads to a costly dead end — while a culture of learning can revitalise an entire ecosystem.
Launched in 2012 in partnership with IBM, the "Oncology Expert Advisor" project aimed to use Watson to provide personalised treatment recommendations to oncologists. After four years of development and $62 million spent, the project was abandoned in 2016 without a single patient ever having been treated on the basis of its recommendations — IBM itself had contractually prohibited their clinical use. Internal documents revealed by investigative journalism flagged recommendations described by physicians as "dangerous and incorrect". The cause was not a simple bug: Watson had been trained on data from the old medical records system (ClinicStation), never updated to integrate with the Epic system adopted by the hospital in 2013. The tool was answering a clinical reality that no longer existed.
The R6 analysis reveals a critical breakdown along the coordination and realisation axes:
Strategic level (S). The strategy rested on a fantasised vision of direct production (S3a): the institution believed that acquiring a technical asset would be sufficient to create medical value. It neglected governance through integration (S2b). The real stakeholders — frontline physicians, nurses, IT systems managers — were not involved in the design of the system, producing a solution disconnected from clinical needs and the existing infrastructure.
Organisational level (O). The project suffered from a major deficit in structuring cooperation (O2b). IBM's engineers and MD Anderson's medical teams worked in hermetically sealed silos, with no organisational mechanism to translate biological complexity and clinical intuition into computational rules. The absence of transformation capacity (O1b) prevented the system from learning from its errors in real time: fed on outdated data, it continued producing recommendations without ever being confronted with operational reality.
Individual level (I). Developers focused on direct execution (I3a) — producing code, algorithms, interfaces — without developing the collaboration competency (I2b) needed to understand the profession they were supposed to be equipping. This disconnection produced a technological hallucination: the tool answered questions nobody was asking, while ignoring vital safety protocols.
In 2014, Microsoft's culture was dominated by "Stack Ranking": a bell-curve evaluation system that forced managers to rank their employees, with a fixed percentage required to be declared "underperforming" regardless of the actual quality of the work. This created a toxic environment of internal competition where the objective was to prove one was the brightest — and, if necessary, to undermine one's colleagues. Satya Nadella identified this cultural blockage as the primary brake on innovation.
The R6 analysis reveals a spectacular pivot along the posture axis:
Strategic level (S). Nadella shifted the organisation from the pole of stability and rent-extraction (S1a) — defending the Windows empire and the dogma of technical superiority — to the pole of adaptation (S1b). He imposed the vision of the "Learn-it-all", accepting that the company did not know everything and had to adapt humbly to market needs and the reality of the Cloud.
Organisational level (O). To anchor this strategy in reality, Microsoft modified its regulatory mechanisms. The company dismantled the competitive evaluation system and replaced it with processes fostering transformation (O1b). The new rules stopped punishing error and began rewarding the "Growth Mindset" — creating an environment where experimentation is structurally encouraged and made safe.
Individual level (I). The most radical change occurred in the very definition of expected competence. Previously, only the brilliant solitary expert (I3a) was rewarded. From then on, individual evaluation explicitly incorporated collaboration (I2b) through a central question: "How have you contributed to the success of others?" This behavioural reorientation unlocked energies, broke down silos, and enabled the massive, coordinated transition to Azure.
The juxtaposition of these two cases illustrates the primacy of S-O-I alignment over raw technological power.
The Watson case demonstrates that the world's most advanced technology cannot compensate for a failure of human architecture. By seeking to "produce" health (S3a) without "structuring cooperation" (O2b) between computational knowledge and medical knowledge, the project built a digital ivory tower. It is a failure by isolation: the technology operated in a vacuum, disconnected from the reality of the field and unable to adjust to the evolution of the hospital's infrastructure.
Conversely, Microsoft proves that economic performance — the share price increased tenfold under Nadella — is a consequence of organisational fluidity. By moving from a culture of rigid accountability (I1a), where error is a fault and knowledge is a weapon, to a culture of adaptive innovation (I1b), where error is a learning data point, Microsoft reactivated its capacity to create value. It is a success by integration: the adaptation strategy (S1b) was only possible because it was supported by individual behaviours of collaboration (I2b) and learning.
Systemic alignment is the true engine of any lasting transformation. You cannot buy artificial intelligence the way you buy a piece of furniture: it requires a human infrastructure of cooperation (O2b) to operate safely.
The lesson is clear: before investing heavily in the tool, invest in your system's capacity to learn and cooperate. A mediocre technology inside an organisation that learns quickly and collaborates well will always outperform a perfect technology inside an organisation paralysed by fear or silos. The true "high tech" is the maturity of your human interactions.