Bayer is a global life sciences company operated through pharmaceuticals, consumer health and crop science. In 2024, the group reported sales of 46.6 billion euros and 94,081 employees.
The company has invested heavily in research, with over 6 billion euros allocated to R&D in 2024, and its leadership frames AI as an enabler for both sustainable agriculture and patient-centered medicine. Bayer's own materials highlight the role of AI in planning and analysis of clinical trials, as well as the accelerated discovery pipeline of crop protection.
In this article, we will explore two mature, internally used applications that convey the central role AI plays in Bayer's core business goals.
- Discovery of herbicides in crop science: Apply AI to narrow down molecular candidates and identify new modes of action.
- Clinical trial analysis of pharmaceuticals: Inges heterogeneous testing and device data to accelerate compliant analyses.
Discovery of AI-supported herbicides
Weed resistance is to increase global challenges. Farmers in the US and Brazil face resistant species to multiple herbicide classes, reducing costs and threatening crop yields. The discovery of traditional herbicides is slow, often from concept to market, 12 to 15 years, expensive, and high exhaustion during early screening.
Bayer's Crop Science Division has turned to AI to shorten these timelines. Independent Report Note Bayer's pipeline includes Icaforine, the first new herbicide action mode in decades, due to be launched in Brazil in 2028, with AI being used upstream to accelerate the discovery of new action modes.
Reuters reports that Bayer's approach uses AI to match weed protein structures with candidate molecules and compresses the early discovery funnels by triating millions of possibilities against pre-determined criteria. Bayer's Cropkey overview describes a profile-driven approach in which candidate molecules are designed to meet safety, efficacy and environmental requirements from the start.
The company claims that Cropkey has already identified more than 30 potential molecular targets and more than 10 potential molecular targets. These figures are promising, but remain assertive until independent verification.
For Bayer Discovery Scientists, AI-guided triage changes the workflow.
- Reduced early stage wet love cycles By focusing on a higher probability of agreement between proteins and molecules.
- Integrating safety and environmental standards Exclude compounds that are unlikely to meet regulatory thresholds on digital screens.
- Moving promising molecules fasterenabling previous testing and potentially compress development timelines from 15 to 10 years.
Reports by both Reuters and the Wall Street Journal are expected to reduce the decline and accelerate the timeline from discovery to commercialization.
The Cropkey program is covered by multiple independent outlets and is a signal of maturity beyond a single press release. Reuters We report Bayer's claim that AI tripled the number of new action modes identified in early studies compared to 10 years ago.
Future squidforin herbicides, which are expected to be commercially released in 2028, indicate that Cropkey's production is expanding into the regulatory pipeline. The presence of both media scrutiny and short-term launch candidates suggests that it is one of Bayer's most advanced AI deployments.
A video explaining Bayer's crop key process in the discovery of crop protection. (sauce: Bayer))
Bayer demonstrates how machine learning can trim low-value screening cycles by focusing AI on high-roy bottlenecks in research and development, and advances only the most promising candidates to experimental tests. At the same time, the acceleration figures reported by the company should be treated as claims until they are supported by multiple seasons, regions, and independent court cases.
Clinical Trial Analysis Platform (Alyce)
Drug development is increasingly dependent on complex data streams. Electronic Health Records (EHR), site-based case report forms, patient-report results, and telemetry from wearables in distributed trials. This data volume and variety controls tensions in traditional data warehouses and slows down regulatory reporting.
Bayer has developed Alyce (an advanced analytics platform for clinical data environments) to handle this complexity. In the Phuse Conference presentation, Bayer engineers will describe the platform as a way to ingest diverse data, ensure governance and deliver analytics more quickly while maintaining compliance.
The presentation explains that Alyce's architecture uses a layered “bronze/silver/gold” data lake approach. The example trial payload contains approximately 300,000 files (1.6 TB) for 80 patients, requiring error handling before standardizing time zone harmony, device ID mapping, and data to SDTM (Study Data Aggregation Model) format. Automatic pipelines provide systematics, quarantine checks, and notifications. These technical details were publicly presented to peers to enhance reliability beyond internal marketing.
For statisticians and clinical programmers, Alice argues:
- Standardize intake Beyond structured (CRF), semi-structured (EHR extract), and unstructured (device telemetry) sources.
- Automate quality checks Free staff to focus on analysis via a pipeline that reduces manual intervention.
- Enable previous insights Reduce delays between data collection and review by preparing analytics-enabled datasets faster.
These objectives are consistent with Bayer's broader statement that AI is being used to safely and efficiently plan and analyze clinical trials.
Phuse is a respected industry forum where sponsors share methods with peers, and Bayer's willingness to disclose technical details indicates that Alyce is in production. Although Bayer has not released accurate cycle time savings, the emphasis on elastic storage, preparation for adjustment and speed suggests measurable efficiency improvements.
Given the specificity of the presentation of real-world payloads, architectural diagrams, and validation process, Alyce appears to be a mature platform that actively supports Bayer's clinical trial program.
Screenshot of Bayer's Phuse presentation showing Alyce's automated ELTL pipeline.
(sauce: Phuse))
Bayer's commitment to Alice reflects broader efforts to modernize and expand clinical development. By integrating different data streams into a single automated environment, the company is located to shorten research timelines, reduce operational overhead, and accelerate promising treatment moves from discovery to patients. The infrastructure also prepares Bayer to expand AI-driven analytics in additional therapeutic areas, supporting long-term competitiveness in highly regulated industries.
While Bayer has not published any quantified cost savings directly linked to specific cycle time reductions or Alice, the company's willingness to present detailed payload volumes and pipeline architectures in Phuse shows that the platform is actively deployed and undergoing peer-level scrutiny. Based on these disclosures and similarities with other Pharma AI implementations, reasonable expectations will improve faster data review cycles, previous anomaly detection, and compliance preparation. Although these results have not yet been publicly tested, Alyce's disparity is restructuring Bayer's trial workflow in ways that could result in significant long-term returns.
