Anthropic PBC has entered the scientific research automation space with the introduction of Claude Science, a sophisticated software platform designed to lighten the workload of researchers conducting experiments in biology and chemistry. The tool, rolling out on June 30, represents a significant shift in how Anthropic is positioning itself beyond general-purpose AI assistance and into highly specialized professional domains where automation can deliver tangible productivity gains.
The core innovation behind Claude Science lies in its integration of more than 60 scientific databases into a unified interface. Rather than forcing researchers to manually query multiple sources of information and piece together results, the platform allows scientists to frame questions in plain language and receive comprehensive answers drawn from integrated knowledge repositories. This architectural approach addresses a genuine pain point in modern research workflows, where accessing and synthesizing information from disparate databases consumes considerable time and intellectual energy. By automating multistep tasks that typically require jumping between systems, Claude Science aims to free researchers to focus on interpretation and hypothesis development rather than data retrieval logistics.
Anthropicunveiled Claude Science during a high-profile event in San Francisco where chief executive Dario Amodei shared the stage with prominent figures from the pharmaceutical industry, signalling the company's ambitions to reshape how drug discovery and development proceeds. Vas Narasimhan, chief executive of Novartis AG and an Anthropic board member, participated in the announcement, alongside Chris Boerner, CEO of Bristol-Myers Squibb Co. The pharmaceutical industry's active engagement with this initiative underscores genuine interest from established players in leveraging AI to accelerate research cycles, though it also hints at underlying concerns about competitiveness and market disruption.
Anthropic is offering Claude Science initially as a beta release exclusively to its paying subscribers, a strategic decision that allows the company to gather detailed feedback from engaged users while maintaining quality control over early deployments. The tool harnesses existing widely available Claude models, specifically Opus 4.8, which Anthropic released in May. Importantly, the company has designed Claude Science to provide traceable details alongside all outputs, enabling scientists to verify information accuracy and understand the reasoning behind generated responses. Similarly, images produced by the application include documentation of the generation process, addressing legitimate concerns within the scientific community about transparency and reproducibility of AI-generated content.
Beyond the software release, Anthropic announced a more ambitious initiative: launching its own preclinical drug discovery programmes conducted entirely in-house. Eric Kauderer-Abrams, the company's head of life sciences, articulated a strategic focus on therapeutic areas that large pharmaceutical and biotech companies might overlook because they fall outside traditional commercial attractiveness thresholds. This represents a fundamental shift from providing tools to scientists toward actually conducting drug discovery research itself, positioning Anthropic as a competitor in the pharmaceutical innovation landscape rather than merely a technology supplier.
Dario Amodei indicated that within the coming year, the company hopes to demonstrate concrete success in using AI to identify novel drug discovery targets. This timeline reflects both genuine optimism about AI capabilities and the urgent need for tangible results to justify the extraordinary valuations commanding the AI sector. Anthropic, currently valued at US$965 billion (RM3.94 trillion), is reportedly preparing for an initial public offering as soon as autumn, making the delivery of measurable real-world outcomes increasingly critical to shareholder narratives.
The competitive dynamics between Anthropic and rival OpenAI have driven much of this innovation push. Throughout the past year, both companies have aggressively developed specialized AI tools across finance, law, healthcare, and now scientific research, motivated by desires to court institutional business customers and justify their substantial capital valuations. This arms race for professional applications has created genuine nervousness across multiple industries about which services and professional roles might become redundant through automation. When Anthropic introduced Claude Cowork in February, its capabilities for automating legal work such as contract review triggered a US$1 trillion (RM4.08 trillion) stock market correction, illustrating how concentrated AI's disruptive potential has become in market psychology.
Vas Narasimhan's comments at the event struck a notably cautionary tone relative to the promotional framing. While welcoming technological advancement, the Novartis chief executive emphasized that appropriate regulatory frameworks must be established proactively rather than reactively following a crisis. His remarks reflect legitimate concerns within established pharmaceutical leadership about ensuring responsible AI deployment in drug discovery, where errors or biases could have profound consequences for patient safety and public health. Narasimhan noted that despite bold proclamations about AI's potential, Anthropic and the broader industry must now deliver genuine clinical benefits to justify the excitement.
The timing of Claude Science's announcement carries particular significance given recent complications with Anthropic's regulatory standing. Less than two weeks before the June 30 launch, the startup had disabled access to its most advanced models, Fable 5 and Mythos 5, following a Trump administration order restricting access by foreign nationals. On June 26, Anthropic successfully obtained US government approval to restore partial access to Mythos 5 after addressing national security concerns, though no resolution has been announced regarding Fable 5. These restrictions highlight the geopolitical complexity now surrounding advanced AI capabilities and the regulatory constraints that companies like Anthropic must navigate when deploying cutting-edge technology.
For Malaysian and Southeast Asian readers, Claude Science's emergence holds several implications. The region's growing biotech and pharmaceutical sectors, particularly in Singapore and Malaysia, may find the tool valuable for accelerating research productivity and reducing operational costs associated with manual data management. However, the automation of scientific tasks also raises questions about whether regional research capacity will evolve toward higher-value interpretation and innovation or potentially create dependencies on Western AI platforms. Furthermore, the regulatory uncertainty surrounding advanced AI deployment suggests that Southeast Asian governments should begin developing clear frameworks for responsible AI adoption in critical sectors like healthcare and drug discovery.
