Stanford Finds Professors Prefer AI Answers as Cuba Ties AI Future to China

Stanford Law School researchers found that law professors preferred answers drafted by AI over those written by their peers 75 percent of the time, according to Complete AI Training, using Google's Gemini 2.5 Pro and NotebookLM to respond to typical first-year contracts questions. Evaluators flagged less than 4 percent of the AI-generated answers as harmful to student learning, compared with 12 percent of the professor-written answers.

The University of Pinar del Rio in Cuba signed a cooperation agreement with the Chinese firm Shanghai Loop Artificial Intelligence Technology to advance AI training, academic mobility and doctoral programs, Cuba Headlines reported, part of a broader Cuban push toward what it calls technological sovereignty. The report notes the effort faces steep obstacles, including an energy crisis that forced temporary university closures in 2026 and unstable connectivity.

IT Brief New Zealand argued that observability is becoming the reliability layer underpinning autonomous AI operations, framing monitoring and instrumentation as prerequisites for moving AI systems toward greater autonomy.

Sources:

• Complete Ai Training
• Cubaheadlines
• It Brief New Zealand