Anthropic Blocks Scientists Tied to States in AI-Assisted Pathogen Research
- Anthropic published a threat intelligence report on Thursday detailing the disruption of attempts to use Claude for biological weapons research, alongside cyberattacks, propaganda, and conventional weapons development.
- The company documented five cases where researchers attempted to use Claude for work with potential biological weapons implications, including research intended for a military research institute discovered in May 2026.
- Beyond biological threats, Anthropic identified actors using Claude for cyberattacks, surveillance, and propaganda campaigns, demonstrating how AI accessibility now enables less skilled individuals to execute sophisticated operations.
- Anthropic banned the involved accounts, shared intelligence with authorities and industry partners, and tightened safeguards, though it cannot concretely demonstrate the users intended to cause harm.
- Microbiologist Susan Monarez warns that the same scientific knowledge supporting vaccine development could enable dangerous pathogen engineering, creating a persistent dual-use dilemma for frontier AI safety.
265 Articles
265 Articles
Anthropic blocks AI weapon plots after 'catastrophic' risk warning
Anthropic has blocked attempts to use its AI models to help develop biological and conventional weapons, days after its own researchers sounded the alarm over the risks posed by increasingly powerful AI. The Claude-maker uncovered five cases in which researchers used its technology “in ways that could support biological weapons development”, according to a new [...]
Anthropic Blocks Suspected Attempts to Use AI for Biological Weapons Research
Artificial intelligence company Anthropic says it blocked multiple suspicious attempts this year to use its advanced AI models in research that could have supported the development of biological weapons, highlighting growing concerns over the misuse of frontier AI systems. The company disclosed the cases in its latest security and misuse report, saying it suspended the […]
The Anthropic artificial intelligence company has stated that in recent months it has blocked several attempts by researchers to use versions of its Claude software in relation to the possible development of biological weapons. The company, competitor of OpenAI (ChatGPT developer), has however warned that it is difficult to establish the underlying intentions of such requests, since the same research can serve both medical and military purposes.…
Russia has written fake news using artificial intelligence, Iran has attempted to spy, and an unknown military entity has requested help in research related to the mutation of the virus.
Researchers apparently wanted to use AI for projects that could also be relevant for the development of bioweapons. ChatGPT competitor Anthropic undercut several accesses.
Anthropic has revealed new evidence of the risks that accompany the rapid development of AI. The American company revealed that it had to identify and block cases of malicious use of the AI model Claude, including for activities that could support the development of biological weapons. In its first Threat Intelligence report published this year, Anthropic says that from December 2025 to August 2026 it identified and stopped incidents of malici…
Coverage Details
Bias Distribution
- 45% of the sources are Center
![[your]NEWS](/_next/image?url=https%3A%2F%2Fgroundnews.b-cdn.net%2Finterests%2Ffb6dc495f74049f513563c33352175eaa0ecd509.jpg%3Fwidth%3D60&w=128&q=75)
Factuality
To view factuality data please Upgrade to Premium
