Jul 2026· Clinical pharmacology and therapy· Vol 120, pp. 648 - 662· 0 citations· 87 references
Medicine
TL;DR
This review examines what clinical pharmacology scientists must become as these systems enter routine practice and whether this shift improves productivity or introduces new failure modes depends on governance and training investments that the field has not yet made.
Abstract
Drug development productivity has not improved despite five decades of computational advancement, with the probability that a compound entering Phase I achieving regulatory approval remaining near 10%. Each automation wave increased throughput while leaving the interpretive bottleneck intact; scientists continued to formulate questions, evaluate outputs, and make advancement decisions regardless of how fast data accumulated upstream. Agentic AI systems capable of reasoning, planning, and executing multi‐step analyses without continuous human instruction represent the first class of ubiquitous computational tools with the architectural potential to compress this bottleneck, but the implications extend beyond efficiency. As computational systems begin to perform interpretation, execution, and evaluation steps that previously required human judgment, the scientist's contribution shifts from conducting analyses to specifying objectives precise enough for autonomous execution and evaluating recommendations that may be difficult to verify independently. Whether this shift improves aggregate productivity depends on whether autonomous systems address the fundamental causes of clinical failure, including insufficient efficacy, inadequate safety prediction, and poor preclinical translation, rather than merely accelerating the analytical work surrounding them. This review examines what clinical pharmacology scientists must become as these systems enter routine practice. The competencies required for effective orchestration differ from those emphasized in traditional pharmaceutical training, and existing governance structures do not address the failure modes that accompany delegation of scientific judgment to autonomous systems. Whether this shift improves productivity or introduces new failure modes depends on governance and training investments that the field has not yet made.
It is argued that agentic bioinformatics should be assessed through workflow correctness rather than final-answer correctness alone, and the Function--Evidence--Validation (FEV) framework is introduced, which separates demonstrated workflow operations, traceable support for actions and claims, and use-case-specific validation.
Productivity in pharmaceutical R&D continues to fall despite deeper biological insight and steady gains in clinical development operations - a phenomenon termed Eroom's Law. Agentic AI workflows powered by reasoning-trained large language models (LLMs), increasingly described as large reasoning models (LRMs), could potentially dent this trend. Unlike earlier task-specific models, these systems couple multi-step reasoning with the ability to plan, invoke external tools and retrieve authoritative information, enabling them to decompose and execute complex scientific and operational tasks. This review discusses agentic AI applications across the drug development continuum, from target discovery to post-market surveillance, and highlights three near-term use cases: algorithmic drug repurposing, informed consent support and automated drafting of regulatory documents. For each, we outline plausible architectures, the current level of supporting evidence and the principal failure modes that constrain deployment. Realizing these gains, however, requires prospective validation, rigorous human oversight and governance frameworks that align algorithmic outputs with clinical, ethical, legal and regulatory standards. When implemented responsibly, agentic AI could transform human-AI collaboration in biopharma, improving R&D efficiency and accelerating delivery of safer, more-effective therapies.
Sheraz Khan, John G. Samuelsson, Xuemei Cai et al.· Drug Discovery Today· 0 citations
Large language model agents have shown promise in bioinformatics, but most existing systems focus primarily on producing final answers, treating planning, tool use, and code execution as transient interactions. This design is poorly suited to long-horizon bioinformatics tasks, where conclusions must remain connected to the data, computations, and intermediate evidence that support them. We introduce \textbf{Bioinfoysis}, a multi-agent harness that represents each request as a persistent, artifact-grounded analysis run. Bioinfoysis combines global planning with step-wise, evidence-driven replanning: the planner maintains an executable checklist and revises pending steps using structured handoffs returned after each worker execution. These handoffs bind intermediate results to their responsible agent, checklist step, and plan generation, preventing stale evidence from being silently reused after replanning. A controlled runtime validates generated scripts, tables, and figures before they are used in downstream analysis or reporting, while role-specific context, persistent memory, and governed bioinformatics skills support reliable execution over long analysis trajectories. We evaluate Bioinfoysis on BixBench and two question-answering tracks of LAB-Bench 2. On BixBench, Bioinfoysis achieves state-of-the-art accuracy of 82.4\%. Across four underlying language models, Bioinfoysis increases average accuracy from 27.81\% to 64.13\% on SeqQA2 and from 3.13\% to 31.25\% on DbQA2. These results demonstrate that reliable bioinformatics automation depends not only on model capability, but also on the harness that governs planning, execution, memory, and evidence flow. We hope that the emergence of Bioinfoysis will play a driving and leading role in the development of the bioinformatics community. Our demo website can be seen in https://report.bioinfoysis.com/.
Qing-Yang Shao, Xin Zhang, Zhou-Yang Yuan et al.· 0 citations
Agentic artificial intelligence is extending bioinformatics beyond conversational assistance by enabling systems to select tools, execute code, revise analytical plans, and interpret biological data. These capabilities may accelerate research, but they also redistribute decisions that determine whether biological conclusions are valid. We conducted a targeted, structured PubMed search in July 2026 and identified 11 peer-reviewed agentic bioinformatics systems for descriptive review based on predefined eligibility criteria for analytical decision-making, tool or code execution, iterative evaluation, or coordinated agent activity. The evidence base covered single-cell transcriptomics, microbial genomics, cancer genomics, and omics applications, together with methodological literature on reproducibility and biological validation. We examined how current systems report delegated authority, provenance, validation, evidence, abstention, and human oversight. Existing platforms implement safeguards such as sandboxed execution, restricted commands, interaction logs, evidence identifiers, automated checks, critic agents, quality scores, and expert assessment. However, published reports rarely provide a connected account linking the original biological question to samples, reference resources, analytical decisions, computational actions, statistical results, supporting evidence, validation outcomes, and final claims. We distinguish inherited bioinformatics errors, errors amplified through autonomous action, and emergent failures arising from memory, retrieval, tool interaction, or agent coordination. We further propose a multidimensional decision-rights profile, consequence-sensitive validation gates, and a claim-to-evidence provenance architecture organized through the Traceable History of Research Evidence, Agent Actions, and Decisions in Bioinformatics (THREAD-Bio) framework. Illustrative cases show that technically successful execution may still support misleading inference. Trustworthy agentic bioinformatics therefore requires claims to remain reconstructible, challengeable, validated, and proportionate to the evidence.
Drug discovery is undergoing a paradigm shift from isolated artificial intelligence (AI) tools to integrated, closed-loop, agent-driven systems that combine prediction with experimental execution. Recent advances in large language models (LLMs), generative frameworks, and self-driving systems are facilitating the emergence of adaptive, multi-agent ecosystems capable of hypothesis generation, iterative optimisation, and autonomous decision-making. Despite this progress, key challenges, including data bias, limited interpretability, coordination fragility, and regulatory misalignment, constrain the translation into reliable clinical outcomes. Here, the evolving landscape of agentic drug discovery is critically examined, illustrating the transition toward hybrid human AI intelligence, digital twins, and regulation-ready autonomous platforms. Resolving these challenges will determine whether the field remains tool-driven or advances into a truly self-evolving scientific enterprise.
