On November 26, 2024, Cradle announced $73 million in Series B funding, led by IVP. This is a historical account of that announcement, with a clarification of the platform and ownership claims based on public documentation checked October 5, 2026 UTC.
A workflow that still needs experimental results
Cradle described a research loop: use generative models and experimental data to propose protein sequences, then learn from measured results. The announcement did not establish that software replaces laboratory validation.
Cradle presents generative AI as a way to suggest candidates and improve selected properties using a team’s experimental data. Its current website describes generation, result uploads and reports as parts of that workflow.
What the platform and vendor figures establish
For readers familiar with ChatGPT, the useful question is how a suggestion becomes evidence. Ask which properties you can measure, how candidates will be tested and how those results inform the next round.
In the 2024 announcement, Cradle reported customer-project speedups of 1.2–12 times and cost reductions of up to 90%. These are vendor-reported results, not a guaranteed saving or an independently measured result from this article. The release does not support a universal “10 to 20 cycles” baseline.
Cradle’s current licensing page advertises ownership of sequence IP, data and reports, with annual software licenses and no royalty or milestone fees. Its security page says customer data trains project-specific models and remains isolated. These are the vendor’s public commitments; review the applicable software agreement before supplying proprietary data. The public website terms govern website use and do not establish the terms of a negotiated software license.
Compare tools on project evidence
If your shortlist also includes Ginkgo Bioworks or Terray Therapeutics, compare the services and evidence for your particular project. A shared AI label is not a like-for-like performance comparison.
When reading investment material, including the linked Andreessen Horowitz website, separate funding from technical validation. A financing announcement alone does not show that a platform will meet your experimental targets.
For any AI tool in this workflow, define the comparison before the trial: target properties, baseline method, assay conditions and acceptable failures.
A market report forecasts an industry; it does not establish that a particular protein-design project will succeed.
What the 2024 funding was intended to support
Cradle said the funding would expand its wet lab, develop additional protein modalities and properties, and improve its models.
The announced plans also included growing engineering, commercial and operations teams. Those were funding intentions in November 2024, not a claim that every planned expansion has been completed.
Stef van Grieken coverage provides further context for the November 2024 announcement. The question for a prospective customer is whether the evidence fits their own protein, measurements and development constraints.
Cradle named Novo Nordisk among its users in that announcement. A named customer is useful context; it does not guarantee the same outcome for another team.
Questions to ask before a trial
Before evaluating a platform, write down the properties you need to improve and the measurements that would demonstrate an improvement. Request a comparison with a relevant baseline rather than relying on an investment headline.
Also clarify data access, retention, licensing and support in the applicable agreement. This article reviews public sources; it does not report a platform trial, laboratory experiment or independent audit.
Sources
The Kingy Brief
Get The Kingy Brief.
AI changes, original tests and one practical thing to try. Fridays at 09:00 Vancouver time.
Free · Double opt-in · Unsubscribe anytime
Signup help and newsletter schedule
Signup form provided by Beehiiv. After submitting, check your inbox for "Confirm your subscription to The Kingy Brief" and open its confirmation link. Check Spam or Promotions if you cannot find it.
Fridays at 09:00 Vancouver time: source-checked AI changes, original tests and one practical thing to try. The weekly restart begins October 9, 2026. We skip a week when there is not enough verified material. Free. Unsubscribe anytime.