Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
This opinion article argues that the Brazilian Air Force (FAB) should begin, in a controlled and institutionally governed manner, the debate and experimentation with doctrine-trained artificial intelligence (AI) agents, heretermed the Synthetic Aviator. The central problem is not merely technological: it is the risk of institution allag in the face of advancing AI in defense, and the growing difficulty of preserving, scaling, and applying doctrinal knowledge with operational consistency. It is argued that an agent trained on the FAB’s doctrinal corpus, operation al history, and decision-making standards could strengthen mission planning, training, and doctrinal compliance, provided it remains subordinate to human authority and robust governance mechanisms. The proposal presents a hypothetical four-layer training architecture an did entifies threelower-risk priority operation al functions for a pilot project. The limits, risks, and conditions for adoption are acknowledged, including the risk of automating doctrinal biases, the sensitivity of operation al data, and the tension between decision support and command responsibility. It is argued that ignoring this agenda wouldbe as serious a mistake as adopting it without criteria.
No comments yet — start the discussion below.