For enterprise technology leaders accustomed to evaluating AI deployments in supply chain and logistics, a new gallery in downtown Los Angeles offers a different but instructive example of what custom AI models, large-scale data collection, and sustainable cloud infrastructure can achieve. On June 20, Dataland — billed as the first “museum of AI arts” in the world — opened its doors, welcoming more than 10,000 visitors in the first two weeks, according to WIRED.
A Museum Built on Custom AI Models
Dataland is cofounded by artist Refik Anadol and studio partner Efsun Erkılıç. Its centerpiece, Machine Dreams: Rainforest, is an immersive architectural installation that responds to visitors’ movements and biometric data, captured via wearable devices like a smartwatch and a shoulder harness. The experience is driven by Anadol’s Large Nature Model, an AI system built from scratch using natural science archives from institutions such as the Smithsonian.
“For three years, we started from scratch and trained our own AI models, and we worked with our own data sets,” Anadol told WIRED. The team traveled to the Amazon and other rainforests to capture raw material, accumulating 5 petabytes of raw data. This approach contrasts with Silicon Valley’s AI firms, which have faced backlash over unlicensed training data. Dataland sourced its data with the consent and participation of researchers.
From Raw Data to Immersive Experience
Visitors enter a large gallery where ultravivid fusions of nature scenes and computer-chip textures swim across walls and floor in a 40-minute cycle. The effects are partly dependent on the movements of everyone in the room. The shoulder harness emits scents, adding another sensory layer. Anadol describes the installation as “impossible to record what it feels like.”
| Technology Component | Specification |
|---|---|
| AI Model | Large Nature Model (custom-trained) |
| Training Data | 5 petabytes from rainforests, research archives |
| Cloud Infrastructure | Google Cloud (with Google DeepMind low-energy resources) |
| Wearable Devices | Smartwatch, plastic shoulder collar with scent emitter |
| Data Sources | Natural science archives from Smithsonian and other institutions |
Ethical AI and Sustainable Compute
Google DeepMind gave Dataland access to “experimental low-energy” resources, allowing the gallery to run on Google Cloud and maintain “sustainable compute.” Anadol has collaborated with Google since 2016, when he became the first recipient of the Google Artists and Machine Intelligence Artist Residency.
Anadol is aware of the negative perception of AI art, often dismissed as “prompt engineering” or “a bunch of eight-second clips.” He says, “I mean, 100 percent, the majority is right.” Dataland aims to prove there are grand possibilities beyond generative slop. “The reason for Dataland is to understand and explain and explore — and tell the world that there is not only one option,” he said.
For enterprise CTOs evaluating AI projects, Dataland demonstrates the value of investing in custom models trained on ethically sourced, domain-specific data — rather than relying on generic, off-the-shelf models. The integration of wearable IoT devices and real-time biometric data processing also mirrors trends in warehouse automation and logistics visibility. The use of low-energy cloud compute from Google DeepMind highlights how AI infrastructure can be both powerful and environmentally sustainable, a consideration increasingly important for corporate ESG goals.
Sources:
