Claude AI Created Something Anthropic Never Designed
Anthropic just found something inside its Claude models that nobody designed or expected. An internal structure called “J-space” appears to work like a cognitive workspace shared across the model. The July 6 research marks a major step in understanding what actually happens insid
Anthropic just found something inside its Claude models that nobody designed or expected. An internal structure called “J-space” appears to work like a cognitive workspace shared across the model. The July 6 research marks a major step in understanding what actually happens inside large language models.
What the J-Space Inside Claude Actually Is J-space is an internal area where Claude appears to gather and share important information across the model. A simple way to think about it is as a whiteboard inside the AI. When Claude answers a question, solves a puzzle, or follows an instruction, key details seem to appear in this shared space so different parts of the model can use them.
Anthropic found J-space using a research tool called the “J-lens.” The tool helps researchers inspect how information moves inside Claude during a task. It showed that J-space emerged during training by itself.
Anthropic did not design it directly. Follow us on X to get the latest news as it happens. Claude’s J-space closely mirrors what scientists call the “global workspace” in human cognition.
Source: Anthropic The idea is similar to a theory in neuroscience called the “global workspace.” In humans, this describes how the brain makes important information available to different mental processes at once. For example, when someone hears a question, remembers a fact, and decides how to answer, those pieces of information need to come together.
Claude appears to do something similar. Anthropic found that Claude can describe what is inside its J-space when asked. It can also adjust those contents if instructed.
More importantly, when researchers directly changed the J-space, Claude’s answers and task performance changed too. This is fascinating for so many reasons, but I am astonished that this legit just *emerged*it wasn't coded or programmed or anything, it just so happens that it when training a large language model, a small internal "consciously accessible" workspace of sorts emerges…..
Đọc thêm từ Tiền số / Crypto

Can Solana Overtake Ethereum’s Market Cap? 3 Reasons It May
Solana (SOL) and Ethereum (ETH) are two rivaling blockchain networks that have their own strengths and weaknesses. Ethereum…
‘Incentivize democracy’ at the cost of security? Inside Bonk’s $20M exploit
Will whales continue accumulating the memecoin despite the governance exploit?

The Quiet Flippening: USDC Is Beating Tether Where Volume Lives
Visa data shows USDC carrying 70% of adjusted stablecoin volume, double USDT. How Circle won settlement, Tether kept savings, and what breaks the truce.

Self-Custody Has Won the Argument, Now It Has to Work: Trust Wallet CEO (Interview)
Trust Wallet's CEO says the next phase of crypto won't be won by ideology, but by products that make self-custody feel as simple, safe, and obvious as mobile banking.