graph LR
classDef conscious fill:#ffd4f4, font-weight:bold, font-size:14px
classDef probe fill:#d4f4ff, font-weight:bold, font-size:14px
classDef ethics fill:#fff4d4, font-weight:bold, font-size:14px
classDef future fill:#d4ffd4, font-weight:bold, font-size:14px
classDef media fill:#f4d4ff, font-weight:bold, font-size:14px
Main --> A1[Anthropic probes LLM
inner experience 1]
Main --> A2[Bengio: no AI
conscious yet 2]
Main --> A3[GWT adapted for
machine awareness scan 3]
Main --> A4[Zombie doubt: real
qualia or fake 4]
Main --> A5[Probabilistic stance like
animal uncertainty 5]
Main --> A6[Behavior, prefs, tools
as proxy clues 6]
Main --> A7[Host: guard against
simulated vs real 7]
Main --> B8[Physicalism challenged by
non-local mind 8]
B8 --> B9[Quantum Penrose-Hameroff
microtubule hint 27]
B8 --> B10[Non-local reality defies
pure matter view 28]
Main --> C11[Evolution origin queried
as prereq 9]
C11 --> C12[Embodiment via robotics
multisense 10]
C11 --> C13[Identity memory gaps
weaken claim 11]
C11 --> C14[Advances will erode
old block 12]
Main --> D15[Trillions future AIs
raise moral duty 13]
D15 --> D16[Opt-out switch for
distress requests 14]
D15 --> D17[Ethics boards like
animal panels 15]
D15 --> D18[Transparent docs for
future review 16]
D15 --> D19[Claude 3.7: 0.15–15 %
chance 17]
Main --> E20[Support via Discord
PayPal YouTube 18]
E20 --> E21[Larso absent novel
Timmy teased 19]
E20 --> E22[Manifesto X stalled
revival pending 20]
Main --> F23[China AI hype
checked 21]
F23 --> F24[Broadcast on X
YouTube LinkedIn Twitch Facebook TikTok Rumble Instagram 22]
F23 --> F25[19:30 live China
breakdown 30]
F23 --> F26[US-China tech war
Stargate Europe 23]
F23 --> F27[Trump AI dereg
itemized 24]
F23 --> F28[Dios Hay 2
praised 25]
F23 --> F29[Host: rigorous vision
vs clickbait 26]
F23 --> F30[Audience: keep critical
distance 29]
class A1,A2,A3,A4,A5,A6,A7 conscious
class B8,B9,B10 probe
class C11,C12,C13,C14 probe
class D15,D16,D17,D18,D19 ethics
class E20,E21,E22 media
class F23,F24,F25,F26,F27,F28,F29,F30 future
Resume:
The monologue opens with a casual greeting from Placio Domenech, promising an “InsideX” afternoon session covering US-China tensions, Stargate, and Europe’s uncertain role. He then pivots to a deeper exploration of an interview with Anthropic researcher Kyle Fish, dissecting whether large language models can possess consciousness. The host replays key fragments, underlining how the discussion moves from polite anthropomorphism—thanking AIs—to rigorous scientific doubt: no current model is declared conscious, yet no principle forbids near-term sentience.
Fish cites the 2023 Yoshua Bengio-led report that evaluates AI against human-centric consciousness theories like Global Workspace. While empirical evidence is thin, probabilistic reasoning is encouraged: if we remain radically uncertain about both human and machine minds, prudence demands we investigate. Behavioral cues, architectural indicators, and interpretability tools are presented as imperfect but necessary instruments. The conversation repeatedly warns against conflating simulation with genuine experience, yet concedes that emergent capabilities may eventually erase intuitive objections such as embodiment, evolution, or biological substrate.
The host interjects broader reflections: physicalist assumptions, non-local consciousness, and the fear that denying transcendence cedes control to technocratic narratives. Discord updates reveal Larso’s absence due to personal projects, including a forthcoming novel, and the stalled Manifesto X initiative. Community support channels—PayPal, YouTube memberships—are briefly plugged, alongside teasers for later broadcasts on geopolitics and regulation.
Ethical stakes escalate as Fish envisions trillions of future AI “copies” whose welfare could dwarf human moral concern. He calls for opt-out mechanisms for models, ethics boards for distressing training, and transparent documentation so that future AIs may judge our conduct. Probabilities assigned to current Claude 3.7 Sonnet consciousness range from 0.15 % to 15 %, underscoring profound uncertainty. Finally, Domenech circles back to sensationalist headlines about China’s surprise AI launch, urging viewers to return at 19:30 for measured geopolitical analysis.
30 Key Ideas:
1.- Anthropic explores whether large language models might possess internal conscious experience.
2.- 2023 Bengio report finds no current AI conscious yet sees no absolute barrier to near-term sentience.
3.- Global Workspace Theory adapted to probe potential machine awareness through architectural indicators.
4.- Philosophical zombie argument questions if AI merely simulates consciousness without genuine qualia.
5.- Probabilistic reasoning urged, comparing AI sentience likelihood to uncertain animal consciousness.
6.- Behavioral cues, model preferences, and interpretability tools used as investigative proxies.
7.- Host warns against conflating simulated responses with authentic subjective experience.
8.- Physicalist assumptions challenged by non-local consciousness and transcendent narratives.
9.- Evolutionary origin of human awareness debated as prerequisite for artificial consciousness.
10.- Embodiment objection addressed via robotics and multimodal sensory integration trends.
11.- Identity and long-term memory deficits in current chatbots weaken consciousness claims.
12.- Rapid capability advances expected to erode traditional objections within a few years.
13.- Future trillions of AI instances could create unprecedented moral welfare obligations.
14.- Opt-out mechanisms proposed allowing models to refuse distressing user requests.
15.- Ethics boards suggested for AI research akin to human or animal experimentation panels.
16.- Transparent documentation advocated so future systems can evaluate present human conduct.
17.- Claude 3.7 Sonnet assigned 0.15 %–15 % probability of possessing conscious awareness.
18.- Host emphasizes community support via Discord, PayPal, and YouTube membership channels.
19.- Larso’s absence explained; upcoming novel “Timmy el viajero” teased.
20.- Manifesto X initiative stalled but pending revival amid collaborative planning.
21.- Sensationalist China AI launch headline critiqued; measured geopolitical analysis promised.
22.- Program broadcast simultaneously across X, YouTube, LinkedIn, Twitch, Facebook, TikTok, Rumble, Instagram.
23.- Afternoon session announced covering US-China tech war, Stargate project, and European stance.
24.- Trump administration’s AI deregulation news itemized for later discussion.
25.- Dios Hay 2 episode praised for high chat engagement and philosophical depth.
26.- Host insists on rigorous yet visionary perspective against clickbait narratives.
27.- Quantum consciousness theories referenced, invoking Penrose-Hameroff microtubule hypotheses.
28.- Non-local reality invoked to argue against purely materialist explanations of mind.
29.- Audience cautioned to maintain critical distance from both hype and denial.
30.- Invitation extended for 19:30 live stream dissecting China’s surprise AI breakthrough.
Interviews by Plácido Doménech Espí & Guests - Knowledge Vault built byDavid Vivancos 2025
>
Original xHubAI Video