Today in Silicon Valley, a drama worthy of a dystopian script is unfolding. Anthropic's reputation as the 'most ethical' artificial intelligence laboratory is undergoing the toughest scrutiny in its history. While #OpenAI and $XAI Elon Musk, according to insider reports, have already accepted military terms, Anthropic has found itself at the center of a scandal due to three words in its contract with the Pentagon: 'any lawful use'.
What is really hidden behind bureaucratic language?
At first glance, the phrase sounds standard. But in the context of defense orders, it becomes a legal foundation for creating autonomous lethal systems. This means that AI will gain the right:
Independently choose and identify targets on the battlefield.
Make decisions about opening fire without direct confirmation from a human operator.
Use algorithms for total digital surveillance in real time.
Importantly: Anthropic has always positioned itself as creators of 'safe' AI with its own 'Constitution', which explicitly prohibits the use of technologies to cause harm. Now this constitution is facing the reality of multi-billion dollar government contracts.
The battle for the startup's future
The situation around the contract has already gone beyond the negotiating rooms. Classic pressure methods have come into play:
Anonymous 'leaks': Officials hint through the press that the company's inflexibility could threaten national security.
Public quarrels: Social networks are buzzing with discussions between supporters of 'clean ethics' and advocates of U.S. military superiority over China.
Financial blackmail: For a startup that requires immense computing power, refusing a Pentagon check could mean losing in the technological race.
Is a compromise possible?
The question remains open: can Anthropic negotiate special conditions that limit the use of their Claude model to logistics or intelligence analysis? Or will the 'right to kill' become a mandatory entry ticket to the club of major players in the AI market?
