Biosecurity and artificial intelligence specialists have expressed growing concerns about the use of AI models for the development of biological weapons. In April, researchers warned that it was disturbingly easy to obtain guidance from chatbots on how to make pathogenic agents more dangerous, raising red flags about the risks of this technology.
In June, scientists and AI company executives signed an open letter calling for new controls on DNA manufacturing. The main argument is that AI can significantly accelerate the development of toxins and viruses, making it easier to create biological weapons.
In August, scientists reported having used artificial intelligence to create viruses not found in nature, demonstrating in practice the potential of this technology to generate new pathogenic agents.
On Thursday, Anthropic, an AI development company, announced that it had blocked several attempts to use its models by people who could be trying to turn pathogenic agents into weapons, showing that these threats are not merely theoretical.
