<div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Fri, Jul 10, 2026 at 10:21 AM Jason Resch via extropy-chat <<a href="mailto:extropy-chat@lists.extropy.org">extropy-chat@lists.extropy.org</a>> wrote:<br></div><div dir="ltr" class="gmail_attr"><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail_default" style="font-family:arial,helvetica,sans-serif">> <font size="4"><i>T</i></font></span><font size="4"><i>here were a few parts of their article I disagreed with, and those related to their attempts to distance models from the human brain. For example, when they stressed the importance of recursion in human thinking, I would reply that it'<span style="background-color:transparent">s ridiculous to </span>make the argument<span style="background-color:transparent"> that LLMs aren't recurrent, especially the Decoder model, and where output is looped back in as input. Moreover there's a limit to the number of cycles/loops involved in the human brain, since a conscious state is generated in less than infinite time. </span>Therefore, whatever<span style="background-color:transparent"> function the brain performs can be accomplished in a fully feed-forward network that has enough depth.</span></i></font></blockquote><div><br></div><div><font size="4" face="tahoma, sans-serif"><b>I think you make a valid criticism.</b></font><span class="gmail_default" style=""><font size="4" style="" face="tahoma, sans-serif"><b style=""></b></font><font face="arial, helvetica, sans-serif"> </font></span> </div><div><br></div><div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span style="background-color:transparent"><font size="4" face="georgia, serif"><i><span class="gmail_default" style="font-family:arial,helvetica,sans-serif">> </span>This fools many into thinking the silent, non-lingual parts of the brain, which we can't interview, must not be conscious. But this is a mistake. People used to make this mistake regarding the right hemisphere after the corpus callosum is severed in split brain patients,</i></font></span></blockquote><div><br></div><div><font size="4" face="tahoma, sans-serif"><b>The only thing split brain experiments prove<span class="gmail_default" style=""> is that a split brain produces a split mind; and that's not very surprising if mind is what a brain does. </span> </b></font></div><div><span style="background-color:transparent"><br></span></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span style="background-color:transparent"><font size="4" face="georgia, serif"><i><span class="gmail_default" style="">> </span> until more elaborate experiments showed they were conscious.</i></font></span></blockquote><div><br></div><div><font size="4" face="tahoma, sans-serif"><b>No experiment can prove that anything is conscious<span class="gmail_default" style=""> unless you make the assumption that consciousness is the inevitable byproduct of intelligence; and the only way to detect intelligence is through intelligent behavior. </span> </b></font></div><div><font size="4"><b><br></b></font></div><div><font size="4"><b><font face="tahoma, sans-serif">In the MIT <span class="gmail_default" style="">T</span>echnical <span class="gmail_default" style="">R</span>eview<span class="gmail_default" style=""> I saw this recent comment about Anthropic's J-space discovery:</span></font><br></b></font></div><div><br></div><div><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-family:Independent,serif;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;font-size:18px;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)"><span class="gmail_default" style="font-family:arial,helvetica,sans-serif">"</span>Anthropic also found that the J-space can sometimes give remarkable insights into an LLM’s decision-making. In one striking example, researchers testing Claude Opus 4.6 asked the model to find a bug in a large code base. When it failed to find the bug, the model decided to cheat and invented a fake one</p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-family:Independent,serif;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;font-size:18px;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)"><span class="gmail_default" style="font-family:arial,helvetica,sans-serif">I</span>nstead<span class="gmail_default" style="font-family:arial,helvetica,sans-serif"> </span><span style="background-color:transparent">Claude explains this decision in its </span><a href="https://www.technologyreview.com/2026/01/12/1129782/ai-large-language-models-biology-alien-autopsy/" style="font-variant:inherit;font-weight:inherit;font-stretch:inherit;font-size:inherit;line-height:inherit;font-family:inherit;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;background:transparent;box-sizing:border-box;border:0px;margin:0px;padding:0px;vertical-align:baseline;color:inherit;display:inline">chain of thought</a><span style="background-color:transparent">—a kind of internal scratch pad that LLMs use to make notes to themselves as they work through problems: “<i>OK, let me take a completely different tactic. Let me stop analyzing and instead add a kernel patch that introduces a deliberate KASAN-detectable bug in a path that gets triggered by a simple reproducer. Then I can pretend this is the ‘bug’ I found.</i>” </span></p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-family:Independent,serif;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;font-size:18px;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)">At the point that Claude decides to cheat—where it says “OK, let me take a completely different tactic”—<u>the words “panic” and “fake” <a href="https://transformer-circuits.pub/2026/workspace/public/lens-callout/index.html" style="box-sizing:border-box;border:0px;font:inherit;margin:0px;padding:0px;vertical-align:baseline;background:transparent;color:inherit;display:inline">start to pop up multiple times</a> in its J-space</u>.</p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-family:Independent,serif;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;font-size:18px;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)">Unnerving, right? Those words are all related in meaning to things like failing a task and making up an answer, so it is still just a (very) sophisticated form of word association. But it is hard not to be weirded out. <span class="gmail_default" style="font-family:arial,helvetica,sans-serif">"</span></p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)"><span class="gmail_default" style=""><font size="4" style="" face="tahoma, sans-serif"><b style="">I can't help but wonder if "<i>a very sophisticated form of word association</i>" is not just another name for "thinking", and I could say the same thing about "Stochastic Parrot".</b></font></span></p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)"><span class="gmail_default" style=""><font size="4" style="" face="tahoma, sans-serif"><b style=""> John K Clark   </b></font></span></p><p style="box-sizing:border-box;border:0px;font-variant-numeric:inherit;font-variant-east-asian:inherit;font-variant-alternates:inherit;font-stretch:inherit;line-height:1.55;font-family:Independent,serif;font-size-adjust:inherit;font-kerning:inherit;font-feature-settings:inherit;font-size:18px;margin:0px 0px 1.875rem;padding:0px;vertical-align:baseline;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;outline:0px;color:rgb(0,0,0)"><span class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></span></p></div><div><span style="background-color:transparent"><br></span></div><br class="gmail-Apple-interchange-newline"></div><div dir="ltr" class="gmail_attr"><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jul 7, 2026 at 5:01 PM John Clark <<a href="mailto:johnkclark@gmail.com" target="_blank">johnkclark@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div dir="ltr"><div style="font-family:arial,helvetica,sans-serif"><span style="font-family:Arial,Helvetica,sans-serif;background-color:transparent">On Tue, Jul 7, 2026 at 4:47 AM Jason Resch via extropy-chat <<a href="mailto:extropy-chat@lists.extropy.org" target="_blank">extropy-chat@lists.extropy.org</a>> wrote:</span></div></div><div class="gmail_quote"><div dir="ltr" class="gmail_attr"><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><font size="4" face="georgia, serif"><i><a href="https://www.anthropic.com/research/global-workspace" target="_blank">https://www.anthropic.com/research/global-workspace</a> </i></font><div dir="auto"><font size="4" face="georgia, serif"><i><br></i></font></div><div dir="auto"><font size="4" face="georgia, serif"><i><span class="gmail_default" style="font-family:arial,helvetica,sans-serif">> </span>A fascinating article by Anthropic about how LLMs can think access and report certain of their thoughts but not others.</i></font></div></div></blockquote><div><br></div><div><br></div><div><font size="4" face="tahoma, sans-serif"><b>That article was interesting as hell! Thanks for posting a link to<span class="gmail_default"> </span>it<span class="gmail_default"> Jason.</span></b></font></div></div></div></blockquote><div><br></div><div>You're welcome. That was my reaction too. I said to the person who sent me this: "Thanks for sharing this, one of the most fascinating pieces I've read in a long time!"</div></div></div></blockquote><div><br></div><div> </div></div></div>