Preserving cultural heritage: Inside Google DeepMind’s collaboration with Pelé

Ashwani Kumar — Thu, 25 Jun 2026 11:07:46 +0000

Preserving cultural heritage: Inside Google DeepMind’s collaboration with Pelé

On August 2, 1959, Pelé scored the most beautiful goal of his career: three consecutive “sombreros” without the ball touching the ground. Yet, the legendary ‘Gol da Rua Javari’ was never caught on film.

At Cannes Lions, we shared the behind-the-scenes story of how we worked together with historians, sports journalists, football legends, and Pelé’s family to reconstruct this lost moment and share it with the world. The work was created in full partnership with Pelé Brand, the official managers of the Pelé estate.

Rooted in education and cultural preservation, the project heads to the Pelé Museum this year. To achieve this, teams from Google DeepMind used frontier models like Gemini Omni and Veo 3 to transform historical fragments into moving imagery, filmed on the original pitch with authentic uniforms and a vintage ball.

By making the memory of this goal accessible, we ensure it stands as a tribute to inspire a new generation of fans during the first World Cup without him.

Watch the full story on the Google YouTube channel soon.

At the Cannes Lions International Festival of Creativity, Google DeepMind unveiled a collaborative project to reconstruct Pelé’s self-proclaimed “most beautiful goal”—the legendary Gol da Rua Javari scored on August 2, 1959—which was never caught on film. [1, 2]

The initiative showcases how advanced generative AI can bridge historical archival gaps to preserve cultural and sports heritage. [1, 3, 4, 5]

The Missing Piece of Football History

During a match against Juventus-SP in 1959, a 18-year-old Pelé scored a spectacular goal by executing three consecutive “sombreros” (flicking the ball over a defender’s head) and a final header without the ball ever touching the ground. Because no television cameras or video recordings captured the moment, the feat lived on exclusively through radio broadcasts, written stadium reports, and local folklore. [1, 6, 7] Preserving cultural

How the AI Reconstruction Worked

Rather than relying purely on machine speculation, the project prioritised historical accuracy by building a vast collaborative network before generating any footage. [6]

The Human Network: Google DeepMind partnered directly with Pelé Brand (the official estate managers), sports journalists, historians, football legends, and Pelé’s immediate family to map out the technical parameters of the play. [1, 6]
The Frontier Models: Technologists used Google’s advanced multimodal and video generation AI models, specifically Gemini Omni and Veo 3, to convert historical text fragments and memories into high-fidelity moving images. [1, 5]
Physical Stunt Conditioning: Modern professional footballers acted as stunt players on the original Rua Javari pitch. They wore authentic, era-specific uniforms and used a vintage match ball. Their physical movements conditioned the AI models, ensuring the generated physics matched real athletic constraints while remaining stylised by Pelé’s archival playing footage. [5, 6]

Creative Leadership and Global Destination

The project was led by KK Walker, Executive Creative Director at Google AI & Gemini. It serves as a visual tribute to inspire football fans following Pelé’s passing. Following its public unveiling, the reconstructed digital media and its accompanying educational assets are heading to a permanent home at the Pelé Museum. [1, 5]