CluePoints Ships Three AI Led Clinical Data Review Capabilities, Co-Innovated With a Top 10 Pharma 🔗

CluePoints, a provider of risk based quality management and clinical data oversight software, made three capabilities generally available: Intelligent Query Detection, Medical and Safety Review, and Intelligent Medical Coding. All three were built and tested inside a top 10 pharmaceutical company's live portfolio and are running across the majority of that partner's ongoing studies, including mega-studies. CluePoints reports that Intelligent Medical Coding targets up to 99% accuracy on MedDRA coding suggestions and roughly halves manual coding effort. The partner joins CluePoints on stage at SCDM 2026 in September.

Weave Bio and Takeda Deploy AI Native Regulatory Automation Across 14 Programmes 🔗

Weave Bio, an AI native regulatory automation company, is deploying its platform across 14 of Takeda's active drug development programmes in multiple global markets. The work began with investigational new drug preparation and has extended to Health Authority Questions and responses. A paper co-authored by both teams reported a 97% reduction in the time to draft an initial submission in one evaluated use case, from roughly 100 hours to three, a self reported figure from a single case. Regulatory experts stay in control while the platform handles drafting, extraction, source traceability and review support.

Genedata Deploys Its Biologics Platform Across Abcam's Global R&D 🔗

Genedata, a provider of enterprise software for biopharma R&D, is rolling out Genedata Biologics across Abcam's R&D sites in the UK, US and China, creating one connected data foundation for antibody development. Both are Danaher companies. The framing is groundwork rather than a single AI product: a unified, structured data layer is the precondition for AI enabled discovery, and Abcam brings roughly 500 new antibodies to market a year onto it.

Veeva Posts a 928 Million Dollar Quarter and Points Its Growth at Agents 🔗
Veeva reported fiscal 2027 second quarter revenue of 928 million dollars, up 18% year on year, with non-GAAP operating income of 415.9 million dollars at a 44.8% margin. It called it the best CRM quarter in its history, with 12 of the top 20 biopharmas now committed to Vault CRM. Veeva Falcon, its agentic platform for clinical, regulatory and safety work, has five early adopters and first go-lives due this year. CEO Peter Gassner said AI is opening the next chapter for Veeva and life sciences.

The signal: Veeva is the closest thing life sciences software has to a bellwether. Strong numbers mean pharma IT budgets are flowing, and the growth is now tied explicitly to agents moving into clinical, regulatory and safety work. That is the same layer IQVIA, CluePoints and Weave Bio are all building into. When the bellwether and the specialists point at the same workflows, the direction is no longer in doubt.

Anthropic Publishes a Standard to Wire AI Agents Into Lab Instruments 🔗

In late August Anthropic released a research preview of the Model Hardware Standard, a common interface between AI agents and the physical instruments that run experiments, from microscopes and liquid handlers to robotic arms and sensors. It is meant to close the part of the loop that leaves the computer: carrying an agent's instructions into equipment and returning measurements for the agent to interpret. Early support was named from AWS, Automata, MBF Bioscience, QIAGEN, Tecan, Danaher, Doosan Robotics and Universal Robots. The deployments are early, require programmable equipment, and Anthropic notes that the models' physical reasoning still needs expert oversight.

The signal: For a year the AI in the lab story has been about planning and analysis. This is about execution: a standard way for agents to actually operate instruments. The life sciences anchors are the tell, QIAGEN on nucleic acid purification, Tecan on liquid handling, Danaher exploring autonomous labs. If a standard like this takes hold, the question for instrument and automation vendors shifts from whether their kit is smart to whether it speaks to an agent.

Adaptyv Raises 40 Million Dollars for Wet Labs That AI Can Run 🔗

Adaptyv, a Lausanne company, raised a 40 million dollar Series A led by Highland Europe to scale an automated wet lab that validates AI designed proteins. A design goes in as a digital sequence; Adaptyv makes the DNA, expresses the protein, measures binding, and returns experimental data through a web platform or API. It grew lab throughput more than fivefold in a year and lists Roche, Novo Nordisk, Chai Discovery and Boltz among more than 100 customers. It recently showed a loop with Anthropic in which Claude designed proteins, sent them to the lab, and received real data back.

The signal: Most of the money in AI drug discovery has gone into better models and more compute. Adaptyv is betting on the opposite bottleneck: models can now propose proteins faster than any lab can test them. The capital says investors increasingly agree that automated wet lab capacity, not model quality, is where the next constraint sits. It is the physical other half of the loop that Anthropic's Model Hardware Standard is trying to standardise.