In a world where humanity put a man on the moon before adding wheels to luggage, the rapid advancements in AI seem almost paradoxical. Microsoft’s chief data scientist Juan Lavista, in a recent Global Stage conversation with Tony Maciulis, highlighted this contrast to emphasize how swiftly AI has evolved, particularly in the last few years.

Lavista discussed the impact of generative AI, which allows users to create hyper-realistic images, videos, and audio. This capability is both impressive and concerning, as demonstrated in their “Real or Not?” Quiz, where even experts struggle to distinguish between AI-generated and real images.

While AI offers incredible tools for good, Lavista warns of the potential risks, particularly with deepfakes and other deceptive technologies. He stresses the importance of public education and the need for AI models to be trained on diverse data to avoid biases.

As AI continues to evolve, its impact on daily life will only grow. Lavista predicts more accurate and less error-prone models in the future, underscoring the need to balance innovation with responsible use.

Watch more interviews from Global Stage.

More For You

- YouTube

At the 62nd Munich Security Conference in Munich, GZERO’s Tony Maciulis spoke with Benedikt Franke, Vice Chairman and CEO of the Munich Security Conference, to discuss whether the post-1945 global order is under strain or already unraveling.

TOKYO, JAPAN - FEBRUARY 8: Japan's Prime Minister Sanae Takaichi, leader of the ruling Liberal Democratic Party (LDP), places a red paper rose on the name of an elected candidate at the LDP headquarters on general election day on February 08, 2026 in Tokyo, Japan. Voters across the country headed to polls today as Japan's Lower House election was held.
Photo by Kim Kyung-Hoon - Pool/Getty Images

When Japanese Prime Minister Sanae Takaichi called snap elections last month, it was a big gamble. Holding a winter election just four months into her tenure with no real policy record to run on?