The Biological Computing Co. (TBC) Releases Proof of Concept for Neurally-Optimized AI Software

The Biological Computing Co. (TBC) Releases Proof of Concept for Neurally-Optimized AI Software

PR Newswire

Interactive OASIS 500M demo reflects progress toward TBC’s first commercial optimized video model, planned for later this year

SAN FRANCISCO, July 30, 2026 /PRNewswire/ — The Biological Computing Co. (TBC), an applied biological computing company that uses real neurons to improve AI models, today released a public proof of concept for its neurally-optimized software.

Based in San Francisco, The Biological Computing Co. (TBC) is harnessing evolutionary intelligence to redefine computing. Our platform integrates living neurons with advanced AI, creating frontier models that are more stable, scalable, and dramatically more efficient for applications including AI processing and real-time adaptive learning.
www.tbc.co

The demo applies TBC’s optimization technology to Decart’s OASIS 500M, an interactive video world model trained on Minecraft gameplay. The playable experience is available at TBC.co.

TBC selected OASIS because the model’s smaller size enabled the team to move quickly and test its approach in an accessible, interactive environment. As part of the proof-of-concept process, TBC worked with AI infrastructure provider Bluesky Compute to conduct matched external testing and host the playable experience on its GPU infrastructure.

Bluesky specializes in high-performance inference operations, its depth in model serving, GPU orchestration, and inference cost optimization. Combined with its ability to support rapid, controlled computational testing, this made it a strong partner for validating the model outside TBC’s internal environment. Under those testing conditions, TBC’s optimized OASIS model demonstrated a twofold improvement on the selected video-quality benchmark, required approximately 4.4 times less inference cost and produced more than three times as much coherent video as the base model.

This demo marks the first application of TBC’s biologically derived optimization principles to a working model, providing an early proof point for the capabilities of its platform. Building on this foundation, TBC is now applying the same approach to a significantly larger, state-of-the-art text-to-video model that will serve as the company’s first commercial offering later this year.

“Finding more efficient ways to run AI is urgent right now, and open-source models are the ideal proving ground: they’re closing the gap with closed frontier systems, and they’re open to modification with new techniques,” said TBC co-founder and CEO Alex Ksendzovsky. “Until today, the industry optimized models with the same shared toolkit, while neuroscience offers a wealth of entirely new techniques. This is the initial proof that principles derived from the brain, the most efficient computer in the world, can deliver real gains for applied AI.”

From real neurons to deployable software

TBC’s Algorithm Discovery Platform brings together biology, computational neuroscience, AI research and software engineering to translate insights from real neural activity into new AI algorithms.

The process begins with an existing AI model and a specific performance problem. TBC’s researchers design targeted experiments on living neural networks, record how the neurons respond and identify potentially useful computational principles within those responses. The team then translates those principles into lightweight software that can be integrated into the original model.

The resulting optimized model runs on conventional GPU and cloud infrastructure. Real neurons are used during the discovery process; users do not need biological hardware to use the deployed software.

Progress toward TBC’s first commercial offering

The OASIS demo provides an early public demonstration of this process before TBC scales the technology to larger, production-grade models. TBC plans to launch its first commercially available, neurally-optimized text-to-video model later this year.

Those interested in participating in the beta program can visit www.tbc.co/early-access.

About The Biological Computing Co.

Based in San Francisco, The Biological Computing Co. (TBC) is building AI systems derived from biology, not just inspired by it. TBC studies how real neurons process information and translates those insights into software that makes modern AI models better, faster and more efficient. Founded by neurosurgeon-scientists and built by an interdisciplinary team of biologists, neuroscientists, engineers and AI researchers, TBC is building the future of biological computing.

Cision View original content to download multimedia:https://www.prnewswire.com/news-releases/the-biological-computing-co-tbc-releases-proof-of-concept-for-neurally-optimized-ai-software-302838521.html

SOURCE The Biological Computing Co.