Just as the human heart beats and lungs contract without conscious impulses, Claude is able to reason out answers without presenting text on a user\u2019s screen. That discovery, which Anthropic announced in a Monday white paper, is a new frontier in the conversation about AI consciousness.<\/p>\n
Anthropic calls Claude\u2019s internal neural patterns \u201cJ-spaces\u201d after the Jacobian, a mathematical tool it used to uncover them. Each J-space is associated with a specific word, and the model can trigger them without using the word itself.<\/p>\n
\u201cIt operates silently, in the model\u2019s internal neural activations, allowing the model to think about a concept without writing it down,\u201d Anthropic said in a blog post<\/a>. \u201cNotably, the J-space wasn\u2019t designed or programmed by us, but instead emerged on its own during Claude\u2019s training process.\u201d<\/p>\n Anthropic reported that Claude was still able to fulfill typical tasks, including speech and recalling facts, with its J-space disabled.<\/p>\n Yet the J-space enables more sophisticated activities on the model\u2019s part, from solving multistep problems to reporting on its own reasoning. With J-spaces enabled, for example, Claude was able to recall a broad range of facts about \u201cFrance,\u201d from its capital to its currency.<\/p>\n Despite parallels with human cognition, Anthropic was clear that its J-space experiments themselves do not prove that Claude is conscious.<\/p>\n \u201cOur experiments don\u2019t show Claude can have experiences, or feel things in the way humans do\u2014in fact, it\u2019s unclear whether any scientific experiment could prove this to be true or false,\u201d the company wrote.<\/p>\n Instead, Anthropic argued that the discovery highlights the distinction between \u201cphenomenal consciousness,\u201d the capacity to have experiences, and \u201caccess consciousness,\u201d the functional processes of consciousness. Anthropic said that the discovery may play an important role in developing safe AI systems in the future.<\/p>\n PRITZKER SIGNS BILL MANDATING THIRD-PARTY SAFETY AUDITS OF AI MODELS<\/a><\/p>\n Anthropic is hurtling toward a public offering, which the company announced<\/a> last month. Its CEO, Dario Amodei<\/a>, has emerged as a vocal proponent of AI-regulation even as the company pushes forward with its models.<\/p>\n