El coste de equivocarse reordena las habilidades profesionales

The cost of making mistakes reshapes professional skills as companies prioritize personnel capable of monitoring algorithms, managing changes, and responding to critical failures. Businesses are now rewarding human oversight over speed, given the economic impact of potential mistakes. Governance and change management have moved up the E&E Barometer, while the cost of allowing machines to operate unchecked increases.

A single financial error, leaked data, or an underused application could negate promised savings. Knowledge about avoiding such issues is growing in importance. A single AI agent in New Zealand can handle financial reconciliations, but not be held responsible for them – that responsibility falls to the financial director. Each case has a corporate responsible, who can authorize decisions, detect errors, and correct them.

The MIT CISR highlights the challenges in separating routine, automated decisions from higher-risk ones requiring human oversight. Only 40% of affected companies modify their policies after an AI incident, while 35% already use AI agents in production. The IAPP identifies eight families of AI governance, including risk, technology, audit, law, and product.

Governing AI involves more than just signing a document of principles; it encompasses inventories, permissions, records, and mechanisms to stop operations. Cybersecurity warns against easy conclusions, as 89% of surveyed organizations experienced AI-facilitated attacks in the past year, with 35% reporting significant financial or operational consequences.

Despite a decline in overall cybersecurity job postings, specific AI security engineering offers surged by 268.3%. The rise of change management and process optimization over traditional project management highlights the shift in focus. While these positions don't measure exact hiring, they suggest that implementing tools is becoming a different issue from improving work performance.

A project may meet budget and schedule but fail due to unnecessary controls or transferring costs between departments. KPMG finds that 79% of surveyed CEOs consider AI a priority investment, while 71% expect a return within one to three years. The pressure creates relevance for skills previously secondary, such as understanding costs, designing processes, managing resistance, and knowing how to suspend an implementation whose results are not sustainable.

Even so, analytical thinking remains the top priority, as the World Economic Forum notes that 69% of employers surveyed consider it essential. However, it's crucial to distinguish between interest and hiring. KPMG reports that 56% of executives consider skill gaps one of their main business challenges, while Workday found a 13% decrease in mentions of training and development in recent job postings.

The surveys don't represent the same companies, but they raise an uncomfortable question: are we creating internal capacity to control and improve new systems, or are we expecting those who are already occupied to do it?