Anthropic CEO Dario Amodei warns AI models could target wrong people, lack human judgment
The company behind Claude is drawing hard lines on military AI, citing reliability gaps that should concern anyone paying attention to the space.
Share
Add us on Google by Editorial Team May. 26, 2026Anthropic, the AI safety company valued in the tens of billions, is putting its money where its mouth is. CEO Dario Amodei has issued a series of increasingly pointed warnings about the risks of advanced AI systems, culminating in Anthropic formally declining US Department of War contracts that sought fully autonomous weapons.
The core argument is deceptively simple: AI models “cannot be relied upon to exercise the critical judgment” comparable to trained military personnel.
Drawing lines in the sand
On February 26, 2026, Anthropic made its position official by walking away from military contracts involving fully autonomous weapons systems.
AdvertisementAmodei’s concerns extend well beyond the battlefield. In his January 2026 essay titled “The Adolescence of Technology,” he characterized the current phase of AI development as a pivotal moment loaded with geopolitical risk. The essay flagged dangers including propaganda manipulation, strategic misuse by bad actors, and the risk of misaligned AI applications that could spiral beyond their intended purpose.
The Mythos problem
In April 2026, Anthropic launched its Mythos model, which promptly identified tens of thousands of software vulnerabilities across existing systems. Amodei warned that governments and organizations have a window of roughly 6 to 12 months to reinforce their cyber defenses before competing models, including those developed in China, catch up to similar capability levels.
Anthropic itself has reportedly incorporated costly measures to prevent its models from assisting in bioweapon development, another indicator of just how seriously the company takes the dual-use nature of its technology.
What this means for investors
Notably, Amodei has not referenced cryptocurrencies or blockchain technology in any of these warnings, indicating a persistent focus on AI safety and defense.
The cybersecurity dimension is particularly relevant for crypto. If the Mythos model found tens of thousands of vulnerabilities in traditional software, the implications for smart contract security, wallet infrastructure, and exchange platforms are worth considering.
The cybersecurity sector stands to benefit most directly. Organizations scrambling to patch vulnerabilities within Amodei’s 6-to-12-month window will need solutions, and they’ll need them fast.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.
