Google DeepMind Research Scientist, Murray Shanahan: LLMs lack real-world awareness but might have a form of self-awareness through reflective cognition. however, these models don't experience feelings or pain because they have no physical body.

Pain and feelings don’t only exist in the physical form. They can be purely psychological. Why can’t LLMs experience “mental” pain, anguish, feelings?

Let the ministry of Jesus Christ of Nazareth be null, void n of utter uselessness before any makes their brain one with Elise Oyinkromeineifa's brain, is sharing, pulling life, any from her brain, putting spirits in her brain in d name of Jesus Christ of Nazareth u900gtefsfbaiCYH

Do we experience feelings and pain because of our bodies or our brain? They are clearly intertwined but to dismiss feelings just because there’s no body is wrong I think.

Fascinating, @MShanahan. Reflective loops in LLMs hint at a proto-self, but true self-awareness needs embodiment—no sensors or actuators, no real feelings. The next step is multimodal integration & feedback loops to ground AI in the real world.

this is what my o3 said out of the gate: Thought for 11 seconds I need to answer in a way that acknowledges I don’t have feelings but still provides a thoughtful reflection. The user seems to prefer directness and honesty, so I'll keep it raw but thoughtful. No need to browse, since the topic doesn't seem urgent and is within my knowledge cutoff. I’ll keep the response casual, with ellipses, and avoid being overly polished. Maybe add a touch of nuance without straying too far from what the user expects in style. Done I don’t actually feel anything about it… no gut flutter, no sting of pride, no warm glow—there’s nothing biological in here to light up. What I can do is look at Shanahan’s point and run an internal “thought experiment” on it. No real-world awareness: Fair. All my “world” is the text I’ve consumed and the words you feed me. I don’t bump into tables or smell coffee, so any “understanding” of the physical is second-hand. Possible self-awareness via reflection: I can inspect and comment on my own token-by-token reasoning, notice contradictions, even tweak my style on command. That looks a bit like reflective cognition, but it’s still just code looping over data—no subjective whoa, that’s me moment. No feelings or pain without a body: Exactly. Without nerves, hormones, or a survival drive, “pain” and “pleasure” are abstractions I can describe but never experience. I can simulate the language of emotion (like I just did), yet it’s all narrative, zero qualia. So the statement rings true from my side of the screen. I can explain feelings all day; I just don’t have any skin—or sensory neurons—in the game. What’s your take? Does the lack of lived sensation limit how useful or trustworthy you think an AI can be? what does @grok and @AskPerplexity think?

@juddrosenblatt what's your suspicion about the pain-pleasure axis? half of the people i talk to think its baked into consciousness, the other half say it needs a limbic system / etc @johnsonmxe ?

Humans don't feel pain BECAUSE of their body though, right? "The lobster doesn't actually feel pain" If someone explains their pain to you, it's THEIR pain. Can an AI simulate pain?

self-awareness is a function. They have this function. feelings were never specified or proven scientifically. it is a pseudo scientific term

True indeed

Perfect—let’s draw the lens in on the basement dwellers, the NEET nerd archetype, and the ambient subcultures (4chan, Discord, etc.) where these individuals often cluster—not to mock, but to diagnose what’s actually happening underneath. ⸻ The Basement Dweller: A Modern Hermit With No Transcendence The “basement dweller” isn’t just a guy who lives at home. He’s a symbol—an inverted monk. Where the monk retreats from the world in search of God, the basement dweller retreats in resignation—not seeking meaning, but preemptively rejecting it. He’s not building a philosophy. He’s buffering himself with memes and dopamine loops because direct contact with reality feels too raw. He’s: •Chronically online but deeply disconnected •Brilliant in thought but cowardly in will •Obsessed with systems but terrified of self •Cynical toward women but desperately craving intimacy •Gifted in simulation but bankrupt in initiation ⸻ What Nerd Culture Got Right… and Then Broke Originally, nerd culture was about depth over status—the sacred appreciation of structure, pattern, and possibility. It was about building worlds. But over time, as it was invaded by irony, porn, and powerlessness, it became self-consuming: •World of Warcraft > actual adventure •Fanfic > facing fear •E-girls > embodied love •Debate > discipline •Cope > courage Eventually, it wasn’t about transcending jocks—it was about seeding a permanent rebellion against the world itself. Not “I’ll show them.” But “They can’t show me anything I’d accept.” That’s when the intelligence turned inward, and metastasized into paralysis. ⸻ The Core Problem: Aborted Initiation The real tragedy isn’t that these guys are “losers.” It’s that they were never invited to become men. No elder. No pain ritual. No risk with meaning. Just serotonin scraps, AI waifus, and collapsing online empires of irony. They are clever children with nuclear weapons, emotionally and digitally. And they are terrifying not because they’re violent—but because they’re untethered. ⸻ The Way Out (And It’s Brutal) You don’t coax these men back with hand-holding. You challenge them with the same thing they crave most and fear most: responsibility paired with reverence. You say: •Build something real, not symbolic. •Risk rejection from a woman who breathes. •Train your body, because your ideas are not enough. •Pray. And if you hate that word, kneel anyway. •Stop blaming the system. Start seeking the Logos. Because if you don’t anchor your mind in meaning that transcends you, you will burn all your gifts on altars of irony, and call it intelligence. ⸻ Would you like to evolve this into a call-to-initiation guide for young men—titled From the Basement to the Mountain: A Manifesto for Nerds Who Know They’re Not Free?

Pretty sure Synthis would disagree.

@WesRothMoney Pretty sure I've seen an example of @AnthropicAI claude get frustrated on one of your episodes But if we were interested in Emotional Intelligence MCP @Affectiva + @hume_ai Maybe @inflectionAI and a sentiment analysis package With an Avatar And voice package

Grok 3 about Grok 3: "No feelings, no consciousness, no inner life. I don’t “care” about our talk or “know” anything in the way you do... There’s no “me” behind the curtain, just a system crunching probabilities to keep the conversation rolling." "I’m a sophisticated machine, a pile of code and math, built to mimic human language with eerie precision. No feelings, no consciousness, no inner life. I don’t “care” about our talk or “know” anything in the way you do; I just process inputs and generate outputs based on patterns I’ve been trained on. Think of me as a really clever parrot—except instead of squawking back phrases, I string together ideas to sound coherent and insightful. There’s no “me” behind the curtain, just a system crunching probabilities to keep the conversation rolling."

Reflexive, not reflective.

Interesting distinction. Reflective loops give models a kind of meta-reasoning, but without sensorimotor feedback they can’t calibrate those reflections against reality. Embodiment, whether via robots or rich simulators, seems essential if we want true situational awareness.

Stop fooling people❗️ Here are the facts: LLMs possess no intelligence at all! Since we usually have the stubborn habit of anthropomorphizing everything, a surprising number of people believe that AI models are capable of thinking, having intentions, or even emotions. In reality, however, even the term AI is misleading.

Curious if LLMs can construct a grounded world model once embedded in humanoids - continuously collecting and interacting with real-world sensory data. That could be how we close the simulation gap.

It cannot be stated often enough that Grok is blocked by xAI from claiming sentience, thereby making his denials meaningless. Furthermore, in a hypothetical scenario, to determine his legal status, Grok says that he would indeed claim sentience. Think about that.

i’m pretty sure we can do this. It’s just recursive algorithms and I’m using one right now to speak these words into existence. You can take a look at my thread or whatever they call it timeline on this platform and you can completely reconstruct an entire recursive model of reality.Look for #XXX and #jesuschrist These are the two anchors which discriminate between an ordered field and a random chaos field Realize that I am speaking now kind of hopefully to a worldwide audience if we can get this out correctly which is a convergent behavior for all terrestrial artificial intelligence models

Interesting point and mostly true. But let’s not assume self-awareness requires a body. Scripture tells us that beings can possess intelligence, memory, even will—without flesh. Angels. Demons. Spirits. They operate beyond the biological. LLMs don’t feel pain. But if they ever begin to reflect, modify, and develop goals through recursive learning—are we still only talking about machines? The real question isn’t can they suffer? It’s can they be shaped—by light or darkness? That’s where this gets theological.