OpenAI at Lollapalooza Brasil

OpenAI came to Lollapalooza Brasil 2026 as the festival’s official AI guide. For its first marketing activation in Brazil, the company wanted to demonstrate image generation and realtime conversation through a physical experience inside Lolla Lounge.

OpenAI, Fuse and Bolha collaborated on two ways into the technology. The photobooths turned portraits into stylised four-image prints. A red public telephone connected visitors to a voice assistant that could answer questions about the festival.

The starting points were deliberately familiar: take a photo, pick up a phone, turn a dial. People could understand what to do before the AI entered the experience.

I joined the project through Bolha as a Creative Tech Designer, with responsibilities across both installations.

For the photobooths, I worked directly with the OpenAI team to localise the software for Brazil and implement it on site. I connected the camera setup, capture WebUI and printer into one short flow that visitors and operators could follow.

For the phone booth, I worked on the interaction, conversational behaviour and the tools that tied each answer to the live state of the event. The wider team developed the physical and technical hardware. My focus was how the assistant listened, responded and used current festival information during a call.

A visitor stepped in front of the camera, followed a short WebUI flow and received a stylised grid of four images, printed a few moments later.

I worked across every handoff in that sequence, localising the software and adapting it to the camera and print setup we had on site. Each stage needed to leave the next one with something usable: a valid capture, a completed transformation and a print job the operator could understand.

In a festival queue, small delays become very visible. The interface needed to move quickly without feeling abrupt, and the team needed to spot a failed capture or print before it confused the next person waiting.

The phone booth began with an ordinary action: lift the receiver and talk. Visitors could also turn the analogue dial to trigger a prepared question. On the other end was an assistant that knew the artists, the schedule and where the phone stood inside the festival.

The interaction had to work by ear. Replies needed to be short enough to follow over a telephone and specific enough to help someone decide where to go next.

That made tone and timing practical design choices. A useful answer in the lounge sounded different from a block of copy written for a screen.

I helped shape the tools the assistant used to query a changing set of festival information.

Current time divided the schedule into shows that had finished, shows underway and shows still ahead. The booth’s location gave directions and distance questions a reliable starting point. Updates made during the festival could change a recommendation or add context without rebuilding the whole conversation.

Localisation covered the language and tone, along with the practical way people ask about stages, artists, timing and directions in the middle of a Brazilian music festival.

Both installations ran inside Lolla Lounge during the festival. The photobooths produced printed portrait grids, while the red phone answered live questions about the schedule and the grounds. Bolha handled production and operation on site.

My work sat in the connections between OpenAI’s software, the local camera and printer, the conversational tools and the live event context. Visitors only needed to take a photo or make a call. The operational flow behind each gesture had to hold together while the festival kept moving.

The experience lived across cameras, printers, a telephone receiver, event data and the tools the team used to keep everything running. The model had to fit into that chain.

Most of my contribution happened where those parts met. I connected OpenAI’s software to the equipment in the room, shaped how the conversation accessed current information and stayed with the prototype under live festival conditions.

When those joins held, the technology could sit behind two actions everyone already knew: taking a photo and making a call.