Anthropic PBC has stepped into the scientific research arena with the introduction of Claude Science, a software platform intended to ease the burden of routine research work for scientists in biology and chemistry fields. Launching on June 30, the tool represents the company's latest push into professional services automation, following similar ventures into legal technology earlier this year that sent ripples through financial markets concerned about AI-driven job displacement.
Claude Science operates by consolidating access to more than 60 scientific databases into a single interface, allowing researchers to pose questions in plain language rather than navigating multiple specialized databases individually. The platform can handle complex, multi-step research tasks such as protein structure prediction, which traditionally requires scientists to consult various sources, cross-reference information, and manually synthesize findings. By automating these processes, Anthropic aims to free researchers from repetitive work and allow them to focus on higher-order scientific thinking and hypothesis generation.
The platform will initially be available as a beta release exclusively to Anthropic's paying customers, giving the company time to refine its capabilities and gather feedback from actual users before any broader rollout. This measured approach reflects lessons learned from the February launch of a legal automation tool, which triggered significant market volatility due to concerns about the potential obsolescence of certain professional services. By limiting initial access to existing subscribers, Anthropic can better demonstrate concrete value before the tool reaches wider audiences.
Anthropically significant for Malaysia and Southeast Asia's growing biotech sectors, the tool's capabilities directly address pain points in research institutions and pharmaceutical companies operating across the region. Scientists in Malaysian universities and research organizations would benefit from streamlined access to international scientific databases, potentially accelerating research timelines and enabling smaller institutions to compete more effectively with larger, better-resourced counterparts. The ability to query multiple databases simultaneously in plain language particularly suits researchers who may not have dedicated bioinformatics specialists on staff.
Beyond Claude Science, Anthropic announced an ambitious expansion into in-house pharmaceutical development, establishing its own preclinical drug discovery programs. According to Eric Kauderer-Abrams, the company's head of life sciences, these programs will deliberately target therapeutic areas that traditional pharmaceutical and biotech companies have overlooked or deemed commercially unattractive. This strategy positions Anthropic not merely as a software provider but as an active participant in drug development itself, substantially raising the stakes and scope of its life sciences ambitions.
The announcement occurred at an event in San Francisco headlined by Anthropic CEO Dario Amodei, alongside Vas Narasimhan, chief executive of Novartis AG and a member of Anthropic's board, and Chris Boerner, CEO of Bristol-Myers Squibb Co. The presence of these pharmaceutical leaders lends substantial credibility to Anthropic's scientific initiatives and suggests genuine industry interest in collaborating with the AI company on drug discovery challenges.
Anthropic's valuation has soared to US$965 billion (RM3.94 trillion), and the company is reportedly preparing for an initial public offering as soon as this fall. These aggressive ventures into specialized professional domains—legal work, scientific research, drug development—form a crucial part of the company's narrative for justifying its extraordinary valuation. Both Anthropic and rival OpenAI have spent the past year building specialized tools across financial services, healthcare, and scientific domains, attempting to demonstrate broad applicability and substantial commercial potential to justify investor expectations.
Yet this expansion strategy carries genuine risk, as evidenced by the February market reaction to the legal automation tool announcement. That event triggered a US$1 trillion (RM4.08 trillion) stock market decline as investors grappled with the possibility that entire professional service sectors could be disrupted or rendered unnecessary by AI systems. Similar anxieties may emerge as Claude Science gains traction, particularly if the tool proves capable of replacing significant portions of research scientists' work.
Narasimhan pointedly addressed these concerns at the event, emphasizing that the pharmaceutical industry must deliver tangible, patient-facing benefits from AI technology to justify the confidence and capital being invested. His comment that "we need to actually show for patients that we actually are delivering real results" reflects growing pressure on AI companies to demonstrate practical, real-world applications beyond theoretical capabilities. The remark also hints at tension between bold technological claims and the measured, evidence-based approach required in pharmaceutical development and medical practice.
Regarding regulation, Narasimhan expressed hope that appropriate oversight frameworks would be established proactively, warning that allowing AI crises to precipitate regulatory action would be counterproductive. This statement carries significance for Southeast Asia, where regulatory frameworks around AI in healthcare and research remain nascent. Malaysia, Singapore, and other regional countries could benefit from watching how governments and industry leaders internationally navigate these governance questions.
Claude Science leverages existing Claude models, particularly Opus 4.8 released in May, ensuring the platform builds on proven technology rather than unproven systems. A key design feature involves traceability: outputs include detailed information about how conclusions were reached, and any generated images include metadata about their creation process. This transparency mechanism addresses a critical concern in scientific and medical applications, where reproducibility and verifiability remain essential to credibility and regulatory acceptance.
The timing of Claude Science's announcement comes barely two weeks after Anthropic restricted access to its most advanced models, Fable 5 and Mythos 5, due to Trump administration orders limiting technology access for foreign nationals. On June 26, Anthropic received approval to partially restore access to Mythos 5 after addressing national security concerns, though Fable 5 restrictions remain in place. This geopolitical backdrop complicates the international rollout of new Anthropic tools and reflects broader tensions around AI technology transfer and national security that directly affect Southeast Asian researchers and institutions seeking access to cutting-edge AI capabilities.
