{"resourceId":"anthropic-claude-scorpion-neurotoxin-binders-17-of-211-20261020","versions":[{"version":"external-f75c3be19122a6c20be5653485eeb97a028b1f871f5866a8b86b9fdb4df18d86","resource":{"id":"anthropic-claude-scorpion-neurotoxin-binders-17-of-211-20261020","title":"Anthropic: Claude ran a largely autonomous 64-hour protein-design campaign against scorpion neurotoxins; 17 of 211 lab-tested designs met binder criteria and no multitoxin design succeeded","organization":"Anthropic; Adaptyv Bio and Twist Bioscience (contract research organizations); Citrini Research","sector":"AI developer research (biomedical protein design), relevant to university and institute labs","geography":"International (vendor research with U.S. and European testing partners; transferable to academic labs with compute and CRO access)","publishedAt":"20 October 2026","publicationDate":"2026-10-20","eventDate":null,"sourceName":"Anthropic","sourceLabel":"Anthropic research post, \"How Claude designed protein binders against scorpion neurotoxins\", 20 October 2026","sourceUrl":"https://www.anthropic.com/research/claude-scorpion-toxin-binder-design","evidenceClass":"vendor-claim","outcomeClass":"mixed","topics":["developers-agents","infrastructure","knowledge-work"],"finding":"Anthropic reported on 20 October 2026 that an internal research model, working mainly autonomously for 64 hours with four follow-up prompts, used 12 design tools to generate 318,484 protein designs, co-folded 553,052 design-toxin pairs and selected 180 designs; staff added 31 more with its help. Two contract research organizations tested all 211 by surface plasmon resonance. Under strict criteria (hits at both labs, every KD 1 µM or tighter, no off-target signal), 17 designs were binders, each against a single toxin; 11 of 60 single-toxin designs (18%), a rate the post calls typical. None of the 91 five-toxin or other multitoxin designs bound more than one toxin. The two strongest Cn2 binders reached single-digit nanomolar affinity. The campaign used on the order of 10,000 GPU hours. No neutralization or animal testing was done; all designs and data are released.","sledRelevance":"Research institutions are deciding how to use agentic AI in wet-lab discovery. This case shows a realistic shape for it: an agent runs the computational campaign, outside labs verify, and humans steer. It also shows the compute bill (about 10,000 GPU hours, which the post admits most researchers lack) and that harder goals, such as multitoxin binders, failed.","evidence":"Opened the Anthropic research post dated 20 October 2026 and read the full text, including methods, testing criteria, results by toxin, limitations, footnotes on GPU hours and lab assessment, and the two outside expert quotes. Did not open the linked technical report or data release in this run.","architectureImplications":"Agent-run design campaigns combine many design and folding tools with large-scale screening, so they need substantial GPU capacity, orchestration and provenance logging; the post notes a simplified single-agent, single-GPU, 24-hour setup reached comparable in silico scores.","governanceImplications":"Agents emailed sequence orders to a CRO (labelled as written by Claude) at staff request, which raises questions about authorization, ordering controls and biosecurity screening when agents interact with outside labs.","securityPrivacyImplications":"No personal data involved. Biosecurity review of agent-designed proteins and ordering is the relevant control area; the post does not describe screening steps.","caveats":"Vendor-authored account of its own model, with a financial research firm as collaborator. Binding was measured only against recombinant toxins; no in-vitro neutralization or in-vivo efficacy. Most of the 180 selected designs, including the strongest, came only after the third human follow-up, so \"autonomous\" is qualified. The technical report was not inspected here.","streamIds":["research"],"roles":{"sales":"Lead with the verified hit rate, not the autonomy story. Anthropic reports 17 of 211 lab-tested designs met strict binder criteria and none of the multitoxin designs worked, with about 10,000 GPU hours spent. Research VPs and protein-design core leaders want to know what agentic design realistically delivers. A bounded engagement scopes an agent-plus-CRO pilot with clear budgets and checks. Interpretation: this is vendor material without neutralization data, so it cannot support claims of faster drug discovery.","engineering":"Treat this as an orchestration and compute problem. The campaign ran 12 design tools, screened over half a million design-toxin pairs and used about 10,000 GPU hours, though the post says a single-GPU, 24-hour setup reached comparable in silico scores. Pre-sales should size compute, tool licensing, provenance logging and ordering controls before quoting. Interpretation; the technical report was not inspected and lab methods are summarized.","delivery":"Frame delivery as running a supervised design-build-test loop. Work covers defining targets and criteria up front, logging agent decisions, approving every external order, and handling data release. The lab PI owns it with research computing and the biosafety office. Interpretation: acceptance should rest on lab-confirmed hits against preset criteria and compute used per hit, because the source reports binding only."},"roleDetails":{"sales":{"customerProblem":"Labs want to know whether agentic AI can run protein-design campaigns that produce lab-confirmed results at an affordable compute cost.","stakeholders":["Vice president for research","Protein design or structural biology core director","Research computing director","Biosafety officer"],"discoveryQuestions":["What GPU capacity can a design campaign draw on?","Which CRO or in-house assays would validate designs?","Who approves orders an agent prepares for outside labs?"],"valueHypothesis":"A scoped agent-plus-assay pilot may show a lab whether agentic design yields enough confirmed hits per GPU hour to justify scaling.","potentialEngagement":"An eight-week pilot running one agent-steered design campaign against a chosen target, with preset criteria and outsourced binding assays.","unsupportedClaims":["Agents produce therapeutics or antivenoms","Fully autonomous discovery without human steering","Hit rates better than conventional design campaigns"]},"engineering":{"fit":"Fits labs with access to design and folding tools, substantial GPU capacity and assay partners; weak fit for labs without validation capacity.","architectureIntegration":"Agent orchestrating design and folding tools on GPU nodes, with provenance logs, a candidate selection step and an approval gate before external orders.","prerequisites":["GPU allocation sized for screening","Licensed design and folding tools","CRO or in-house binding assays"],"constraints":["Large GPU demand for full campaigns","Biosecurity and ordering policies","Vendor model availability and terms"],"security":"Apply biosecurity screening to designed sequences and require human approval for any order an agent prepares.","proofOfValue":["Compare hit rate against a recent human-led campaign on a similar target","Report GPU hours per lab-confirmed hit","Test whether a reduced single-GPU setup gives comparable in silico scores locally"]},"delivery":{"work":"Define targets and acceptance criteria, run the agent campaign with logging, approve orders, coordinate assays and release data.","dependencies":["Compute allocation","Assay partner agreements","Biosafety review"],"ownership":"Lab PI with research computing and the biosafety office.","skillsAdoption":"Training lab members to steer agents with follow-up prompts and to audit design decisions in logs.","governanceCheckpoints":["Biosafety review before design starts","Human approval of each external order","Data release review"],"acceptanceCriteria":["Proposal: binder criteria fixed before testing","Proposal: GPU hours per confirmed hit recorded","Proposal: every external order traced to an approving person"],"risks":["Compute cost exceeding budget","Overstating autonomy or therapeutic value","Agent-prepared orders bypassing review"]}},"retrievedAt":"2026-10-20T22:05:45-05:00","enrichedAt":"2026-10-20T22:05:45-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Humans still chose the problem, wrote prompts and follow-ups, added designs and ran oversight; lab scientists and CROs remained essential for validation.","procurementImplications":"Pairing agent design with CRO testing implies budget lines for GPU hours and outsourced assays; institutions should plan agreements with CROs that cover agent-generated orders.","operatingModelImplications":"A design-build-test loop split between an AI agent, internal scientists and external labs needs clear ownership of steering decisions, ordering approvals and data release.","sourceVerification":{"openedUrl":"https://www.anthropic.com/research/claude-scorpion-toxin-binder-design","referenceExcerpt":"Of the 211 designs our teams assessed, 17 met our criteria for a binder, each against a single toxin.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}