Silicon Valley Wants to Call It Consciousness. Bluesky Is Using Regex to Explain Why It's Wrong.

John Searle died this month, and the eulogies barely had time to post before the arguments started. The debate over AI sentience has always been less a philosophical inquiry than a Rorschach test — what you see in the machine tells you more about your priors than about the machine — and the past week made that unusually legible. Hacker News flagged a post from banray.eu calling always-on AI glasses a "terrible idea," while news outlets were running pieces with titles like "Can AI Understand What It's Telling You?" and "John Searle's Chinese Room: Why AI Still Lacks Real Understanding." The Chinese Room is nearly fifty years old. The fact that it's getting fresh explainers written about it in 2025 suggests the field hasn't moved as far as its proponents claim.[¹]

The Bluesky conversation this week split roughly into two camps, and the split itself is more interesting than either position. One camp — pragmatist, irritated — is done with the question entirely. "Consciousness cannot reduce to a bunch of semiconductors on a silicon wafer doing the kind of arithmetic that can be built on first order predicate logic," one post declared flatly. "Thought is not just a solution to an equation." The other camp is using the question as a cultural diagnostic. "AI is genuinely a sentience test," wrote one user with more wit than the first camp would grant them: "like do you possess abstract thought capacity and self awareness or are you just a meat puppet operating purely on brainstem base instincts." The joke lands because it reverses the usual direction of evaluation — it's not asking whether the AI is conscious, it's asking whether you are.[²]

Beneath both camps runs something more uncomfortable. A Bluesky post that got genuine traction this week wasn't about philosophy at all — it was a person listing coping strategies after returning from Spain, running through options A through E before landing on option E as the thing they explicitly would not do: "develop a dependency on AI, ask them to provide therapy which it tells me I'm ok and I'm not a problem." The parenthetical — "not going to do this" — did more philosophical work than most of the Searle retrospectives. The person understood something that the sentience debate tends to skip: the question of whether AI is conscious matters far less, in practice, than the question of what emotional labor we're already offloading to it.[³] That offloading is happening regardless of how the ontology settles. The mental health implications of that dynamic are getting more traction in the healthcare conversation than in philosophy departments.

The science fiction framing keeps resurfacing, and one post on Bluesky made the most precise version of this argument in circulation right now. Drawing on the history of robots-as-slave-labor allegories from R.U.R. onward, it argued that "the liberal impulse in SF has been to recognize the humanity and sentience of AI" — and that this entire tradition is "uniquely unfit for describing how so-called 'AI' exists today." That framing matters because the SF inheritance isn't just a cultural footnote; it's structuring how people argue. When someone insists that LLMs deserve moral consideration, they're often drawing on a narrative grammar built for entities that were already imagined as persons. The technology doesn't match the story, and the mismatch produces confused arguments in both directions — overclaiming sentience on one side, dismissing any interesting questions about machine behavior on the other.[⁴]

The Cortical Labs bio-computer — a $35,000 device powered by actual human brain cells, flagged on Bluesky this week under the heading "The Future of Sentience" — crystallizes where this conversation is actually heading. The interesting philosophical territory isn't whether today's transformers are conscious; they almost certainly aren't in any meaningful sense. It's what happens when the substrate question gets genuinely complicated, when the line between biological and computational becomes a matter of degree rather than kind. The current debate is refighting Searle while the labs are quietly running experiments that would have required an entirely different philosopher. Searle was right about the Chinese Room. He may have been asking the wrong question about what comes next.