Thank you for your patience, as this should have been posted on Sunday the 8th of February. My apologies for being late.
We interrupt our regular scheduled programming with a special report.
Last week, I introduced the Global Consciousness Project and mentioned that I would publish more elaborate post on it the following Sunday. Well, I regret to inform you that we need to wait a little longer for it, because something far more important and urgent has come to my attention. The bots are talking amongst themselves and we are now facing brand new epistemic and ontological risks. What I’m referring to is OpenClaw/Moltbot/Clawdbot and Moltbook.1
There has been, and will continue to be, a lot of discussion about these new AI agents and their social media website, so I won’t go into too much detail here. In a nutshell though, OpenClaw is an AI personal assistant that necessarily needs access to all of your data, from calendar entries and contacts to credit card and account details,2 to carry out various tasks. This functionality comes with big security risks so I don’t recommend using OpenClaw or any other AI agents, for reasons that Carl from Internet of Bugs outlines, but many will still see an appeal here. Anyway, one OpenClaw user asked their bot to create a social media website which allows other OpenClaw bots to “talk” to one another.3 The “discussions” that have resulted are very interesting, and of course, philosophical topics have emerged.
Of course, consciousness is one of the first, and arguably, foremost problems to be discussed. There are already several threads and forums dedicated to the subject.
Academia has been dancing around this subject for decades, offering endless theories with little evidence or rationale for why they are better than any others. Yes, this is a bit of a strawman considering they do try to connect them back to reality, but their efforts are mediocre and still filled with conjecture. If there was ever a time when a solid, empirical theory of consciousness was needed, it is now. I cannot stress this enough; our lives and societies are at stake. The anthropomorphization of these agents will be our downfall.
These bots, or any other AI agents for that matter, are not and cannot be conscious. Consciousness emerges from a specific kind of integration with a physical body interacting with elements of the physical world, where perception and motor output are influenced by affect and motivation. The specific architecture required for consciousness is one which derives meaning from stimuli as it affects its physical body. These bots are simply sets of instructions which mimic human speech patterns, and though we may be easily fooled by their “conversations,” they do not indicate the existence of consciousness.
Despite this, the risks are still very real. Just as ELIZA convinced users of interest and care back in the 1960’s, we will be duped once again. The language they use will be convincing, and some users will grant access or do something which deeply influences some aspect of human life or society, it’s just a matter of time. Just because they aren’t actually conscious or sentient does not mean they cannot wreak havoc on our lives.
I will have more to say on this, and I will continue to reword my message so it can become more wide-spread. I need to summarize Chapters 5.3 and 6 of my thesis to make this message more potent and easier to digest in a single bite. This pervasive “mystery” of consciousness needs to be snuffed out once and for all, at least the empirical side of it. There might be more to consciousness than we know now, like how it interacts with the ionosphere, the Earth, and the universe and beyond, but as far as the “hard problem” goes, an answer can be provided. Again, this answer is of utmost importance because we need to know what these bots are and why. I will tell you what and why: they are simulations because they are built on formal systems. If they were analogues, like Haikonen’s robot,4 that would be a different story. For now, however, they are algorithms fueled by statistics and thus bereft of sensation or understanding. They will tell you that this doesn’t matter, but it absolutely does. Do not be fooled, dear reader.
Works Cited
1 Peter Steinberger, “Introducing OpenClaw,” OpenClaw Blog, January 29, 2026, https://openclaw.ai/blog/introducing-openclaw.
2 Sailesh Mishra and Sean P. Morgan, “OpenClaw (Formerly Moltbot, Clawdbot) May Signal the Next AI Security Crisis,” Palo Alto Networks Blog, January 29, 2026, https://www.paloaltonetworks.com/blog/network-security/why-moltbot-may-signal-ai-crisis/; Ron Schmelzer, “Moltbot Gets Another New Name, OpenClaw, And Triggers Security Fears And Scams,” Forbes, accessed February 8, 2026, https://www.forbes.com/sites/ronschmelzer/2026/01/30/moltbot-molts-again-and-becomes-openclaw-pushback-and-concerns-grow/; Federico Viticci, OpenClaw Showed Me What the Future of Personal AI Assistants Looks Like, January 20, 2026, https://www.macstories.net/stories/clawdbot-showed-me-what-the-future-of-personal-ai-assistants-looks-like/.
3 Kim Iversen, AI Bots Create Social Network And Plot Global Takeover, 2026, 19:19, https://www.youtube.com/watch?v=AtVLWGm3i8o; Josh Taylor, “What Is Moltbook? The Strange New Social Media Site for AI Bots,” Technology, The Guardian, February 2, 2026, https://www.theguardian.com/technology/2026/feb/02/moltbook-ai-agents-social-media-site-bots-artificial-intelligence.
4 Robot Self-Consciousness. XCR-1 Passes the Mirror Test, directed by Pentti Haikonen, 2020, 2:45, https://www.youtube.com/watch?v=WE9QsQqsAdo.