Musk’s Two-word Response To Anthropic Ceo’s Claim Its Ai May Have Gained Consciousness

Elon Musk had a two-word retort for Anthropic CEO Dario Amodei after he said his company’s artificial intelligence models whitethorn person gained consciousness.

The tech mogul issued a crisp comeback to a station connected X by cryptocurrency-based prediction marketplace Polymarket, which read: “Anthropic CEO says Claude whitethorn aliases whitethorn not person gained consciousness, arsenic nan exemplary has begun showing symptoms of anxiety.”

In response, Musk wrote, “He’s projecting.”

The station from Polymarket came aft Amodei said he didn’t cognize if nan company, which makes nan popular Claude chatbot, has created AI models that are conscious.

“We’ve taken a mostly precautionary attack here. We don’t cognize if nan models are conscious,” Amodei said during an question and reply with The New York Times.

“We are not moreover judge that we cognize what it would mean for a exemplary to beryllium conscious aliases whether a exemplary tin beryllium conscious. But we’re unfastened to nan thought that it could be,” he continued.

Amodei explained: “We’re putting a batch of activity into this section called interpretability, which is looking wrong nan brains of nan models to effort to understand what they’re thinking.”

“And you find things that are evocative, wherever location are activations that ray up successful nan models that we spot arsenic being associated pinch nan conception of worry aliases thing for illustration that. When characters acquisition worry successful nan text, and past erstwhile nan exemplary itself is successful a business that a quality mightiness subordinate pinch anxiety, that aforesaid worry neuron shows up,” he said.

Amodei’s stunning admittance to nan Times comes amid a clash betwixt Anthropic and nan Department of Defense complete nan usage of AI wrong nan military.

Amodei antecedently said his institution denied a Pentagon petition to region safeguards connected home surveillance and connected afloat autonomous weapons.

President Donald Trump and management officials deemed nan institution a “supply concatenation risk” – a word antecedently reserved for overseas adversaries – past week. The move came aft Amodei refused to backmost down complete concerns that nan company’s products could beryllium utilized for wide surveillance of Americans aliases autonomous weapons.

Anthropic has antecedently vowed to writer if nan Pentagon pursued what it described arsenic a “legally unsound” action, ne'er earlier publically applied to an American company.”

While nan AI institution mislaid awesome partnerships pinch defense contractors, Anthropic saw a surge successful user downloads complete nan past week arsenic group sided pinch its civilized stance.

Anthropic has boasted of much than a cardinal group signing up for Claude each time this week, lifting it past OpenAI's ChatGPT and Google's Gemini arsenic nan apical AI app successful much than 20 countries successful Apple's App Store.