Anthropic has discreetly installed a new biology laboratory in the San Francisco Bay Area, United States. This new facility aims to allow the company to transition from computer-based studies to conducting physical experiments while expanding its projects in life sciences.
The company plans to employ artificial intelligence to streamline investigations and find solutions for rare diseases and conditions that the pharmaceutical sector considers financially unattractive. Part of this work will be conducted internally, while another part will be carried out in collaboration with external partners.
Eric Kauderer-Abrams, who is responsible for life sciences at Anthropic, confirmed the existence of this facility during an interview given to Reuters. He emphasized that practical laboratory trials remain a crucial phase in biological research, stating that, to conduct biology, the final test will continue to be real laboratory work for a period.
An Anthropic spokesperson clarified that the laboratory's main objective is not the discovery of new drugs, and the company chose not to disclose details about the specific activities developed at the site.
Furthermore, the company is working to use Claude to control robotic equipment capable of conducting experiments with minimal human intervention. However, human supervision is still considered indispensable to ensure safety.
In June, Anthropic had announced that it would begin preclinical work in areas not considered profitable by traditional companies. This initiative covers conditions classified as 'undesirable,' for which drug development still presents major difficulties.
One possible approach involves applying AI in research related to bispecific and trispecific antibodies, which are molecules capable of interacting with multiple points on a protein or cell.
Anthropic's expansion includes other significant moves. The company confirmed the acquisition of Coefficient Bio, a deal reportedly valued at approximately US$400 million (about R$2.05 billion), although the company did not comment on the exact value.
This growth occurs at a time when Anthropic itself signals the inherent risks of artificial intelligence. Researchers at the organization recently debated the possibility of the technology assisting in human extinction, and the company also identified scenarios where its systems could be used in the development of biological weapons.
Balancing potential and risk
Kauderer-Abrams emphasized the company's need to balance the great potential of AI to accelerate research with the dangers arising from its application. He explained that in the field of life sciences, they are constantly adjusting this balance: seeking to improve models in ways that drive and accelerate drug creation, while remaining cautious about how these models are implemented to mitigate risks.
Anthropic also needs to manage the concerns of pharmaceutical industry clients. These clients may fear that information about their development programs could fall into the hands of competitors, despite the company assuring that data is kept confidential.
Currently, Anthropic does not conduct clinical trials. The company states that its focus is on research and automation phases, aiming to meet demands little explored by the market, without entering direct competition with pharmaceutical and biotechnology companies responsible for drug commercialization.