Anthropic vows to 'lay risks on the table' while limiting Mythos AI model
Anthropic says it will openly present the risks tied to its new Mythos model even as it limits how the model is deployed because of its powerful cybersecurity capabilities.
Anthropic, the American AI developer, says it will ‘lay the risks out on the table’ concerning its new model, Mythos. The company plans to restrict deployment after acknowledging Mythos’ powerful cybersecurity capabilities, which raise stark questions for companies and governments.