Latent Labs Unveils AI Platform to Make Protein Design Accessible
Latent Labs Unveils AI Platform to Make Protein Design Accessible
Latent Labs, a biotech innovator founded by Simon Kohl—formerly of DeepMind’s AlphaFold protein design team—has launched a new web-based artificial intelligence model named LatentX. This platform aims to democratize the design of novel proteins by making advanced AI tools accessible via a user-friendly web browser interface.
State-of-the-Art Performance in Protein Engineering
Since emerging from stealth mode with $50 million in backing, Latent Labs has focused on building a foundational biology model that pushes the boundaries of what is possible in protein design. According to CEO Simon Kohl, LatentX has "achieved state-of-the-art on different metrics"—meaning it delivers leading performance based on key measurements for protein functionality and viability in lab testing.
Unlike traditional models that primarily predict existing protein structures, LatentX can design entirely new proteins and molecular structures such as nanobodies and antibodies. This capability opens the door to creating tailor-made molecules for novel therapeutics, diagnostic tools, and biotechnologies—potentially accelerating medical breakthroughs and enabling rapid development of targeted treatments.
A New Paradigm: Generating Proteins, Not Just Predicting Them
LatentX distinguishes itself from legacy models like AlphaFold by offering protein generation rather than just structure prediction. As Kohl explains, "AlphaFold is a model for protein structure prediction… but it doesn’t let you generate new proteins." With LatentX, research teams and companies can use natural language prompts to describe the characteristics they want, and the AI generates candidates with atomic-level precision.
Accessible to Academic, Startup, and Pharma Users
Unlike other AI-driven drug discovery startups—such as Xaira, Recursion, and DeepMind spinout Isomorphic Labs, which focus on developing their own proprietary medicines—Latent Labs licenses its platform to external users. This approach allows academic researchers, biotech startups, and pharmaceutical companies to leverage top-tier AI capabilities without building their own in-house systems or assembling complex AI teams.
- No coding required: Protein design can be carried out directly in the browser.
- Free to start: LatentX is initially free, with advanced features planned for future pricing.
- Supports collaboration: Suitable for institutions of any scale.
The Ecosystem and Backing
Latent Labs is joining a growing movement of companies—such as Chai Discovery and EvolutionaryScale—making open foundational AI models available to the life sciences sector. The company is supported by major investors including Radical Ventures, Sofinnova Partners, Google’s Jeff Dean, Anthropic CEO Dario Amodei, and Eleven Labs CEO Mati Staniszewski.
Implications for Drug Discovery and Beyond
The launch of LatentX could significantly reduce the time and resources needed to develop new proteins for medicine, research, and industry. By providing advanced AI tools in a web-based format with natural language controls, Latent Labs aims to lower the barrier to entry and help a broader range of innovators participate in the next generation of biological discovery.
