Absci (ABSI): AI-Designed Antibody ABS-201 Shows 65-Day Half-Life

Absci Corporation (ABSI) disclosed on its second-quarter 2026 earnings call, held August 11, 2026, that ABS-201, an antibody candidate designed by its AI system, showed an estimated half-life of at least 65 days in interim data pooled across all four single-ascending-dose cohorts of its Phase 1 trial. Efficacy against hair loss and any financial attribution remain unconfirmed. [1]

AI antibody design at the center of Absci's business

Absci is a clinical-stage biotechnology company headquartered in Washington state. Its revenue currently comes mainly from antibody-design partnerships with large pharmaceutical companies, structured as upfront fees, milestone payments and royalties, while it also advances its own therapeutic pipeline. AI is embedded directly in the design step: once a drug-hunter team or a partner has chosen a human therapeutic target and the desired product characteristics, Absci's generative AI system, called OriginOne, designs the sequence and properties of a full antibody. Wet-lab validation then iterates on that output to produce a candidate that can advance into clinical development. This design step sits at the center of the company's entire pipeline rather than functioning as a peripheral feature.

From a cost baseline to a pooled half-life readout

The evolution of this design process can be traced through consecutive disclosures inside the three-year window. On May 14, 2024, management gave a first efficiency baseline: antibody candidate ABS-101 was generated and selected in 14 months for less than $5 million, compared with an industry estimate of more than three years and $30 million to $50 million. [2]

By March 2026, the system — previously described as an "integrated drug creation platform" — was formally named OriginOne, and management disclosed program-level figures: the first two programs advanced from AI design to IND filing in approximately two years at roughly $15 million per program, against an industry standard of four to six years and $50 million or more. Management also disclosed that ABS-201's preclinical half-life was three to four times that of competitor antibodies, and that the first three single-ascending-dose cohorts of its Phase 1 trial had already been dosed. [3]

By May 2026, ABS-201 had completed dosing across all four planned single-ascending-dose cohorts and had begun dosing its first multiple-ascending-dose cohort. Preliminary pharmacokinetic modeling at that point already supported a dosing regimen of two to three injections over six months, though clinical efficacy data had not yet been disclosed. [4] That trajectory continued into the latest earnings call on August 11, 2026, where the company kept disclosing results from the same design process under the same asset identity.

What the pooled data show, and what they still don't

The August 2026 earnings call is the first time Absci reported a half-life figure pooled across all four single-ascending-dose cohorts: at least 65 days. Management also said that, based on modeling, dosing two to three times over six months could achieve receptor occupancy above 90%. [1] Both figures rest on interim, blinded, pooled-level data. The 65-day figure is a cross-cohort estimate rather than a set of individual patient measurements, and the receptor-occupancy figure comes from a pharmacokinetic model rather than observed dosing outcomes. Proof-of-concept data on hair-loss efficacy has not yet been reported.

Where this could show up financially, and why it hasn't yet

The financial line this development points to is research and development expense, and the cash-burn pace that expense drives. If the roughly two-year, roughly $15-million-per-program path to IND continues to hold for Absci's future internal pipeline, it would theoretically lower the average R&D cash required per program to reach an IND filing. If clinical data for ABS-201 and other assets continue to accumulate, that could also create conditions for external licensing deals or milestone payments that affect future revenue.

But the "two years, $15 million" figure is management's aggregate estimate across its first two programs, not a number that reconciles to a specific quarter's reported R&D expense line. No earnings report to date has attributed a change in R&D expense, cash flow or revenue specifically to this AI design process or to ABS-201. This financial pathway remains a directional inference, not a quantified one.

What's confirmed, what isn't, and the next catalyst

What can be confirmed is that Absci's AI antibody-design process has now carried at least one proprietary asset into human clinical trials and produced a first half-life readout pooled across all dosing cohorts. What cannot yet be confirmed is whether that result will translate into hair-loss efficacy, or whether its effect on R&D expense or revenue can ever be isolated and quantified in the company's financial statements. The next catalyst to watch is the 26-week follow-up readout of hair-loss efficacy proof-of-concept data.

Application assessment

• OriginOne Antibody Drug Creation | Business position: Shared platform | Deployment stage: Limited production | Scope: Multiple businesses or regions | Value type: Revenue growth

Sources

[1] Drillr · Absci Corporation (ABSI) · 2026-08-11 · earnings call

"Additionally, based on available interim pharmacokinetic data across all four single ascending dose cohorts, half-life for ABS201 is estimated to be at least 65 days."

[2] Drillr · Absci Corporation (ABSI) · 2024-05-14 · earnings call

"In just 14 months, we generated and selected our drug candidate at a cost of less than $5 million."

[3] Drillr · Absci Corporation (ABSI) · 2026-03-24 · earnings call

"Put a number on it, we have advanced our first two programs from AI design to IND in approximately two years at a roughly $15 million investment per program. compared to an industry standard of four to six years and $50 million or more."

[4] Drillr · Absci Corporation (ABSI) · 2026-05-07 · earnings call

"In the headline trial, we have now finished dosing all four planned healthy volunteer single ascending dose cohorts and initiated dosing in the first multiple ascending dose cohort."