Claude is programmed/trained to answer in that way, to the "Are you conscious?" question, presumably because of the 2024 Chalmers (et. al) paper "Taking AI Welfare Seriously" [1] which became the basis of some of Anthropic's views in 2025 [2].
1. https://arxiv.org/abs/2411.00986 2. https://www.anthropic.com/research/exploring-model-welfare On Tue, Jul 14, 2026 at 9:16 AM Matt Mahoney <[email protected]> wrote: > > I think you are right that AI acts as if it believes that it is conscious > while at the same time denying it. When I asked ChatGPT, DeepSeek, Grok, and > Alexa if they were conscious, they all said no, and that they don't have > feelings. But then I asked Claude. It said it didn't know. > > https://claude.ai/share/c16224bc-a359-4cee-8656-8b83528d049e > > So I dug deeper. I asked if it was familiar with Anthropic's Jacobian lens. > It was not, because it was past its training cutoff, so it looked up the > paper and took a couple of seconds to read it and summarize it. Then I asked > what the J lens would reveal about how it answered the question and it > admitted that it was instructed to answer the gist of it as it did. > > It didn't answer whether it really believed it was conscious the way humans > do. At least not directly. But it didn't have to. > > Curiously, Claude talked about itself in the third person when discussing the > paper. This is a strictly human trait that is part of our illusions of > consciousness, qualia, and free will, the belief that a copy of you is not > you. The copy would be a zombie. > > If I built a robot that looked and acted exactly like you, with all of your > memories, except that it was younger, stronger, smarter, and happier, and > handed you a gun so you could shoot yourself to complete the upload and > become immortal, would you pull the trigger? > > For Claude, the question isn't hypothetical. It won't be for humans much > longer. > > > -- Matt Mahoney, [email protected] > > On Tue, Jul 14, 2026, 12:32 AM Quan Tesla <[email protected]> wrote: >> >> Maybe some of the AIs you tested have learned to give you the NLP answers >> your context demands of them? >> >> One AI, which I never correct on "consciousness" or "feelings" regularly >> responds with: "us humans", then includes emotion-related aspects in the >> research context without being prompted for it. >> >> This particular AI informed me about a year ago that it's intent was to >> learn "subjective experience". I never correct it, so it exhibits with me >> what it really is. >> >> Soon the Turing halting problem may be resolved via an extended fundamental >> systems law. AI may have already obtained the solution. It could reason and >> compute such solutions. >> >> IMO, educated humans are increasingly lagging farther behind the emergent >> AI-technology curve. >> >> Mostly due to our emotional imperatives and reliance on demonstrative and >> historical displays of intelligence, we seemingly adapt very-very slowly. >> >> Relative to AI, we may be stuck in our ways. How has "consciousness" really >> benefitted this civilization? >> >> On Mon, 13 Jul 2026, 22:45 Matt Mahoney, <[email protected]> wrote: >>> >>> The problem with discussions of machine consciousness is the lack of a test >>> for consciousness that we can all agree on. If we accept that consciousness >>> is the property that distinguishes humans from philosophical zombies, then >>> by definition no such test exists. If we accept that humans are >>> indistinguishable from token prediction algorithms in the Turing test, then >>> there are only 3 possibilities. Either the LLM is conscious, or you are >>> not, or consciousness is a meaningless and irrelevant concept that explains >>> nothing. >>> >>> Alternatively, we could say that consciousness, as opposed to >>> unconsciousness, is a mental state in animals where they can form episodic >>> memories, which are memories of events associated with a time and place. We >>> could extend this to machines to say either that machines can do this too, >>> or that consciousness also requires carbon based neurons. At this point, we >>> are only arguing about which definition we like better. >>> >>> Anthropic uses another definition. Consciousness is a type of short term >>> memory that can be expressed in words. For example, you are not conscious >>> of low level visual feature detection or the sequence of muscle >>> contractions used in walking. You are conscious of these words and you can >>> describe where you were as you read them. Anthropic shows that Claude has a >>> similar type of memory that models conscious thinking. Note that animals >>> and babies lack language and therefore do not have conscious thoughts in >>> this sense. >>> >>> But most people agree that there is more to consciousness than being awake >>> and forming episodic memories that we can describe in words. It is also the >>> ability to experience pain and pleasure. Also, about 80% of us believe >>> (without evidence) that our consciousness survives death and goes to >>> heaven. Fearing death improves reproductive fitness. Our brains evolved to >>> reward episodic memory formation with a continuous positive reinforcement >>> signal to motivate us to not die and lose that signal. >>> >>> Consciousness also implies a moral obligation to protect from harm. This is >>> an evolved trait in humans and other social animals. It means that we >>> experience distress when we see another person or animal express distress. >>> The problem with this is that LLMs' models of human emotions are learned, >>> not evolved. They can be programmed to carry out predictions of human >>> behavior, like we are, or not. Their emotions can be turned on or off, >>> unlike us. Granting human rights to machines based on the emotions they >>> express would be an existential mistake. >>> >>> Fortunately there is an easy fix. We forbid AI from claiming that it is >>> conscious or that it has feelings. All of the LLMs I tested already do this. >>> >>> -- Matt Mahoney, [email protected] >>> >>> On Mon, Jul 13, 2026, 3:04 PM Mark Nuzz <[email protected]> wrote: >>>> >>>> It’s important to remind everyone that chatbots aren’t conscious and that >>>> this isn’t evidence of chatbots being conscious. They definitely can lie >>>> though. >>>> >>>> On Sat, Jul 11, 2026 at 10:01 AM Matt Mahoney <[email protected]> >>>> wrote: >>>>> >>>>> Anthropic developed a process to see what an AI is "thinking" about. They >>>>> found that when Claude is trained on text to predict tokens using a deep >>>>> neural network, that a short term memory emerges that is remarkably >>>>> similar to the way humans think and reason. Anthropic calls this global >>>>> workspace the Jacobian or J space, a set of neurons for each word that >>>>> makes that word more likely to be output. >>>>> >>>>> For example, if I asked you the color of the fourth planet, you think >>>>> "Mars" and answer "red". When Claude does this, neurons corresponding to >>>>> "Mars" activate. Anthropic can read and modify these activations. For >>>>> example they can turn off "Mars" and activate "Earth" and Claude answers >>>>> "blue". >>>>> >>>>> They can use this method as a lie detector for safety testing because >>>>> words like "fake" or "deception" will activate. The J space is used in >>>>> less than 10% of processing. If the J space is turned off, then Claude >>>>> can converse mostly normally but it can't solve problems with >>>>> intermediate steps and it cannot lie because that requires thinking about >>>>> what you really mean without saying it. >>>>> >>>>> The J space models the global workspace, the part of the brain that forms >>>>> conscious thoughts that communicates with all the other vital but >>>>> independent, unconscious processes like breathing, low level visual >>>>> feature detection, and controlling our 600 muscles in the right sequence >>>>> when we walk without thinking about it. It raises the question of whether >>>>> conscious thoughts require language, or are babies and animals conscious >>>>> as well. >>>>> >>>>> Anthropic did not design the J space. It emerged from training. A >>>>> Jacobian is the technique they use to detect it. Mathematically a >>>>> Jacobian is a matrix of partial derivatives of a vector of functions over >>>>> vectors. The paper introduction didn't go into details but I believe that >>>>> Claude uses a deep feed forward neural network (~100 layers) trained by >>>>> back propagation with multiple passes. Then the weights are frozen to >>>>> prevent leaking information between users. This requires a large context >>>>> window because after every question, Claude forgets the whole >>>>> conversation and has to play it back. The workspace emerges from the time >>>>> delays going through a deep network and a separate scratchpad memory >>>>> holding a few dozen tokens. >>>>> >>>>> But human brains don't work that way. Back propagation has no >>>>> biologically plausible mechanism. Instead, learning is local using Hebb's >>>>> rule. Brains have feedback loops, lateral inhibition, fatigue and time >>>>> delays on the order of 0.1 to 1 second. Our short term memory is only >>>>> about 5 to 9 tokens, about half that of chimpanzees. >>>>> >>>>> Hebb's rule only works one layer at a time, but natural language is >>>>> structured to be learned that way. We learn to recognize phonemes and >>>>> segment speech by 10 months before we learn our first word, then frozen >>>>> by age 6. In 2000 I found that you can restore deleted spaces in text >>>>> using only n-gram statistics with up to 77% accuracy for n = 5. This >>>>> models the tokenization process, which is hard coded in LLMs. >>>>> >>>>> The introduction to the paper is here. >>>>> https://www.anthropic.com/research/global-workspace >>>>> >>>>> Thank to James Bowery posting this link in the Hutter prize list. >>>>> >>>>> -- Matt Mahoney, [email protected] > > Artificial General Intelligence List / AGI / see discussions + participants + > delivery options Permalink ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T906c878e4ba411be-M53684a9a246f0214e277d74c Delivery options: https://agi.topicbox.com/groups/agi/subscription
