Anthropic PBC has introduced Claude Science, a new artificial intelligence platform set for rollout on June 30, designed to help researchers accelerate their work by automating repetitive laboratory and computational tasks. The tool represents the latest in a competitive push by major AI developers to position themselves as indispensable partners across professional sectors, from law and finance to the sciences and healthcare industries.
The software integrates connections to more than 60 scientific databases, creating a unified interface where researchers can query information using natural language rather than navigating multiple specialised platforms. Scientists will be able to automate complex workflows spanning biology and chemistry, with protein structure prediction cited as a key capability. Rather than moving between separate tools and databases, researchers can now pose questions and receive synthesised answers drawn from multiple authoritative sources, potentially accelerating hypothesis development and experimental planning.
Claude Science emerges from Anthropic's growing ambition to become a central infrastructure provider for knowledge work. The company has spent the past year rolling out AI applications targeting specific professional verticals. In February, Anthropic introduced Claude Cowork, a tool automating legal document review and briefing preparation, a move that triggered market anxiety about job displacement. These successive announcements have contributed to broader concerns among investors and professionals about which roles and services AI might ultimately render unnecessary, creating waves of volatility in technology stocks.
Anthropologic has moved beyond simply providing tools for external use. At the San Francisco event announcing Claude Science, the company revealed it is now pursuing its own preclinical drug discovery programs in-house. This represents a strategic shift from being a pure software provider to becoming an active participant in pharmaceutical research itself. Eric Kauderer-Abrams, Anthropic's head of life sciences, indicated the firm would target drug development areas that traditional pharmaceutical and biotech companies have overlooked, potentially opening new markets and therapeutic approaches.
The launch attracted prominent pharmaceutical leaders as speakers and stakeholders. Vas Narasimhan, chief executive of Swiss drug maker Novartis AG and an Anthropic board member, shared the platform with Anthropic CEO Dario Amodei. Bristol-Myers Squibb Co CEO Chris Boerner was also expected to participate. This convergence of AI and pharmaceutical leadership underscores the stakes both industries perceive in this technology's potential applications.
Narasimhan struck a cautious note despite his company's enthusiasm for AI applications. He emphasised that pharmaceutical companies must demonstrate tangible patient benefits rather than making optimistic claims without evidence. More pointedly, he called for regulatory frameworks to be established proactively rather than waiting for crises to force government action. His comments reflect growing awareness in the corporate world that ungoverned AI deployment carries reputational and operational risks.
Claude Science operates on Anthropic's existing Claude models, including Opus 4.8, released in May. A critical feature addresses a persistent weakness in AI systems: transparency and verifiability. The platform generates outputs accompanied by traceable details allowing scientists to verify accuracy and understand the reasoning behind results. Images produced by the system will similarly include metadata explaining how they were generated. This design choice acknowledges that in scientific and medical contexts, blind trust in AI outputs is untenable and potentially dangerous.
Anthropic's valuation has reached US$965 billion, and the company is reportedly preparing for an initial public offering as soon as autumn. These ambitious timeline signals reflect investor confidence but also pressure to demonstrate that lofty valuations rest on genuine commercial viability rather than speculative enthusiasm. Successful deployment of Claude Science in academic and corporate research environments could provide tangible revenue streams and use-case validation that public markets will scrutinise carefully.
The timing of Claude Science's rollout is complicated by recent regulatory developments. In late June, the Trump administration imposed restrictions on access to Anthropic's most advanced models, Fable 5 and Mythos 5, citing national security concerns about foreign nationals acquiring cutting-edge AI technology. Anthropic successfully lobbied for and obtained permission to restore limited access to Mythos 5 after addressing these concerns, but Fable 5 remains restricted. These constraints highlight how geopolitical tensions and artificial intelligence capabilities are becoming intertwined at the policy level.
For Malaysian researchers and the broader Southeast Asian scientific community, Claude Science represents both opportunity and challenge. Universities and research institutions across the region could potentially access more sophisticated research tools at lower cost than maintaining separate subscriptions to premium databases and computational services. However, integration with such tools carries data sovereignty questions that governments and institutions should carefully examine before adoption.
The competitive landscape shaping these developments remains fluid. OpenAI and other major AI firms are simultaneously developing similar professional tools, suggesting a broader shift in how knowledge workers will interact with information and computation. Dario Amodei's stated hope that the coming year will show "some success" in using AI for drug target discovery suggests realistic expectations, at least publicly, about timelines for delivering transformative results.
Anthropic's dual strategy of providing tools to external researchers while pursuing its own drug discovery programs creates interesting dynamics. The company gains insights into pain points and capabilities through customer engagement while potentially positioning itself to commercialise successful discoveries directly. Whether this approach strengthens or conflicts with customer relationships remains to be seen. For now, Claude Science represents Anthropic's bet that automating scientific research will prove as valuable as automating legal or financial work—and that the company itself can evolve beyond being merely a software vendor into a participant in the discoveries its tools enable.
