2026-10-27T08:06:13Z https://nostr.now Ulrike Hahn https://nostr.now/npub1u328aygluds5ft6n3ngrwjmd3ptq9j4xryeek708ryw524krrndqt4xcf5 https://fediscience.org/system/accounts/avatars/109/301/127/580/127/072/original/336fe0d3912e93dd.jpg https://fediscience.org/system/accounts/avatars/109/301/127/580/127/072/original/336fe0d3912e93dd.jpg https://nostr.now/nevent1qqs8klgyz29905988wzc9s8s2f9j80tpjct09l0w2nwrr3ngf287wwczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ux2w3f In reply to <a href='/nevent1qqswnapg49xfpd9u2x0qmv6x2ulxqve6rvqfa8yp0xsnr9d582s0husk3cfen'>nevent1q…cfen</a><br/>_________________________<br/><br/>not just move “where”, but move “with what remaining energy, time, ….” 2026-10-30T10:46:16Z https://nostr.now/nevent1qqstsndw2dsmx6lnww6d90q87fk6xahsvf9xh6v4mwlfaph5fn4c72qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5mrp4s7 trying this one again as previous attempt broke (sorry!)….<br/><br/>poll question pertaining to online social media:<br/><br/>are there any topics where you think online debate has been clarifying, in the sense of leaving participants (collectively) better informed or with clearer understanding?<br/><br/>the question isn’t whether you personally found useful bits of information, but whether you think things ended well overall<br/><br/>I can think of [xxx] examples where I think the collective outcome has clearly been beneficial: 2026-10-29T21:15:45Z https://nostr.now/nevent1qqsvc8gkqqe83z7te30xdh0y6ecl4lfa2t2g4qthk3wy02qnfyhpyxczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rhmgdp In reply to <a href='/nevent1qqsqynjyhxlexcx455ajfqe9mnulayv8pdu878der0tktnw6kw44hmqdy0r46'>nevent1q…0r46</a><br/>_________________________<br/><br/>thinking of open systems 2026-10-29T09:59:51Z https://nostr.now/nevent1qqs96jec24atxkfynxtkhx7jczm2nqwc69wm5fqema3a5ak9hrakv2szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5xxxjtp a poll question pertaining to online social media:<br/><br/>are there any topics where you think online debate has been clarifying, in the sense of leaving participants (collectively) better informed or with clearer understanding?<br/><br/>the question isn’t whether you personally found useful bits of information, but whether you think things ended well overall<br/><br/>I can think of [xxx] examples where I think the collective outcome has clearly been beneficial: 2026-10-29T08:46:33Z https://nostr.now/nevent1qqsqvmqk32n6n0cmale3kphpjg5afktz8rqp0psxlnpyhpyygu7ffzgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5gg4kan report on the state of US incursions into academic freedom<br/><br/><a href="https://www.growkudos.com/projects/from-anecdote-to-evidence">https://www.growkudos.com/projects/from-anecdote-to-evidence</a> 2026-10-28T08:25:09Z https://nostr.now/nevent1qqsptwcqx7c4jkgztu7wfv5qc4asq0getsmrecrahqu49qpvwvhjwlczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd540w7s5 a real threat to digital democratic participation<br/><br/><a href="https://www.theguardian.com/society/2026/oct/27/ai-bots-flooded-2023-uk-government-consultation-disposable-vapes-ban">https://www.theguardian.com/society/2026/oct/27/ai-bots-flooded-2023-uk-government-consultation-disposable-vapes-ban</a> 2026-10-28T07:21:43Z https://nostr.now/nevent1qqsyhsa8xw24x0ax34xhnrtvx72cad0yjvjv3u5nwn76rxdupfplhrgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5jwkuqq now that all the world’s paywalled research has been ingested by large language models, could we maybe at least ditch the need for two factor authentication every time I try to access such academic articles? 2026-10-27T14:56:36Z https://nostr.now/nevent1qqs8egvgl7je9twxueg3qnmwcle8znuyz0awtmtlwzfq3sd894779hczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5nzprrl In reply to <a href='/nevent1qqsrmdhskska90jqffnl89mtrqv9gyntyz7n0mn70k3zt0klnr6g9lq8hr6rw'>nevent1q…r6rw</a><br/>_________________________<br/><br/>koennte man, je nach Sachlage, so sehen 😉…der Text selber sagt natuerlich nichts darueber wie etwaiger Normenkonflikt zwischen dieser und anderen Normen im Einzelfall zu loesen ist… 2026-10-26T16:13:33Z https://nostr.now/nevent1qqs89jfjczwrfxdn3ak83dtpj8xqpuqsf8n95tuywcmrfuet7l4q2xczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5sfvp3d I&#39;ve long been interested in the question of what the social norms are for contributing on large micro-blogging platforms (do people have divergent views on those norms ? is a consistent set of norms is even possible?)<br/><br/>I remember a joke here a while back that caricatured US, German and Balkan users of Mastodon with respect to fundamental intuitions (something like: &#34;This is my thread, bug off&#34;, &#34;showing others they are wrong is a public service&#34;, &#34;we argue with everyone&#34; -respectively).<br/><br/>This new philosophy paper in the journal Synthese articulates essentially that &#34;German&#34; position: we have communitarian obligations toward others&#39; knowledge states<br/><br/><a href="https://link.springer.com/article/10.1007/s11229-026-05830-4">https://link.springer.com/article/10.1007/s11229-026-05830-4</a><br/><br/>#philosophy<br/><br/> #philosophy 2026-10-26T12:34:21Z https://nostr.now/nevent1qqsrxf5zs3ckmyr9zljph9n75mtgxsnpdyej43tlg7yh4dmedgg598gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5e7w5u4 In reply to <a href='/nevent1qqst5jxwhezv9nscc6zgwjdkmq4ymmykj4g0xxxpyn9k7vexsxwjk8cw9awf3'>nevent1q…awf3</a><br/>_________________________<br/><br/>yes, all that, but still surprising to me. The biggest limitation is probably the narrow number and, with that, range of human judges which would likely have pushed the evaluation into more generic “helpfulness” territory and somewhat away from specific research area based assessment (ie “what level of contribution does this make”) 2026-10-24T07:38:39Z https://nostr.now/nevent1qqs0r5ruhkqlswhh08rjsknxw0wkwdkqg6e85m4n8h5fxktcyskkfwczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ena3y4 there’s no “knowledge” without a “knower” and the bounds of human knowledge will remain what humans know… 2026-10-24T07:29:47Z https://nostr.now/nevent1qqsy7fx0wd3n0gs996nfg4s3egskzlauyfadlruds369ffn502kpptqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5yqyrx7 well put, I thought<br/><br/><a href="https://proofsandprompts.com/2026/10/21/artificial-intelligence-and-the-mathematical-community/">https://proofsandprompts.com/2026/10/21/artificial-intelligence-and-the-mathematical-community/</a> 2026-10-24T07:24:16Z https://nostr.now/nevent1qqsyjfhurmr0xlp8he75fsuhchnsafnm9n8thylx590sgyxu2cc4jugzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5u0wcs4 Reinventing mathematics in the era of AI<br/><br/><a href="https://www.nature.com/articles/d41586-026-03389-1">https://www.nature.com/articles/d41586-026-03389-1</a> 2026-10-24T07:09:32Z https://nostr.now/nevent1qqsgya6jcf5n5lprg56thr3cp8k2wcak9864494fckwqsl5vl638mkczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd59qaa4h In reply to <a href='/nevent1qqspjhd47tfprw2ywzf0kvtfeppxf0afgec28jea50ln297psq9j94s3fuf8l'>nevent1q…uf8l</a><br/>_________________________<br/><br/>the level of performance surprised me … 2026-10-24T07:03:28Z https://nostr.now/nevent1qqs2srskl3ysjqw6tcx57qcg2f2u9whvq0yayz66p597w94nzs6dwlqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5f79v0d In reply to <a href='/nevent1qqsxc2vcjer32x7ycgsz30zqer5ft9yn0tvdq4qdck5uyrlt0ushm9ggcw8d5'>nevent1q…w8d5</a><br/>_________________________<br/><br/>that leeway exists, for sure, but I’m doubtful that some individuals not going down that route will be enough to alter the overall direction of travel. Or to put it differently, relying on individuals to resist instead of changing the incentive structure in response doesn’t strike me as sufficient - certainly not for students, and probably not for ECRs either 2026-10-21T18:50:27Z https://nostr.now/nevent1qqsrh4lxczj0upeuzm44dd0fc70n0j2epxxexs92zst4u230t0mx4ugzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5kft9us interesting article on academic freedom by the president of the International Communication Association - expanding his presidential address at the 2026 ICA conference<br/><br/><a href="https://academic.oup.com/joc/advance-article/doi/10.1093/joc/jqag017/8931742">https://academic.oup.com/joc/advance-article/doi/10.1093/joc/jqag017/8931742</a> 2026-10-21T12:11:09Z https://nostr.now/nevent1qqsdp3zmfxllkgfpsfqka0eyncz6eftvekp9qyvwyg3l7yl242maqlqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5g8wgku In reply to <a href='/nevent1qqsq406r63ky4f8vny32cm3l8clsk2pguh55uscygjm4y5g5vktzflgsqlcxj'>nevent1q…lcxj</a><br/>_________________________<br/><br/>if all the short term incentives fail to reward that position, though, then it’s ultimately doomed…<br/><br/>and the current incentives (for both students and researchers) don’t…. 2026-10-21T11:39:55Z https://nostr.now/nevent1qqsdca0fj20xvnrk5t5zj0alk4ve0t8yvcy62cqy8qumanf02uvaz9czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd59kap2a “Grant terminations targeting health research for BIPOC and SGM communities have disproportionately targeted the careers of BIPOC, SGM, and female researchers and researchers from disadvantaged backgrounds. Ensuring the health of all Americans will require systematic re-investment in these topics and scientists”<br/><br/><a href="https://www.thelancet.com/journals/lanam/article/PIIS2667-193X(26)00108-0/fulltext">https://www.thelancet.com/journals/lanam/article/PIIS2667-193X(26)00108-0/fulltext</a> 2026-10-20T18:53:11Z https://nostr.now/nevent1qqs2zkyl233vs93jk0kte7tm6jkla89kh4ne6x45jwhl8s74anpzwnszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ky65tm “And once all students have fully adopted the voice of AI, we’ll know we’ve finally experienced educational equality.”<br/><br/><a href="https://www.mcsweeneys.net/articles/i-need-ai-to-write-better-lesson-plans-so-my-students-stop-using-ai-to-write-their-papers">https://www.mcsweeneys.net/articles/i-need-ai-to-write-better-lesson-plans-so-my-students-stop-using-ai-to-write-their-papers</a> 2026-10-20T17:30:16Z https://nostr.now/nevent1qqs9x797t0dtxfecjtmut3yagpmj494j290dcywdxpm2xppzqqqe8wqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5072te6 In reply to <a href='/nevent1qqsy0yx0yxc604w7kpt427elf5dtgug35rqa4xpemj567vhae9rve6cw8697r'>nevent1q…697r</a><br/>_________________________<br/><br/>#academia 2026-10-20T15:29:44Z https://nostr.now/nevent1qqsy0yx0yxc604w7kpt427elf5dtgug35rqa4xpemj567vhae9rve6czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5748fcg interesting study into the effects online harassment of scientists has on other, “bystander” scientists<br/><br/><a href="https://journals.sagepub.com/doi/10.1177/10755470261481083">https://journals.sagepub.com/doi/10.1177/10755470261481083</a> 2026-10-20T15:29:44Z https://nostr.now/nevent1qqstkpr5zfensxnks3d8dvr339yng0xwph04yethlkhatyy953l7rzczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5esmanc new Dutch code of research conduct, including new passages on the use of genAI<br/><br/><a href="https://www.knaw.nl/en/publications/netherlands-code-conduct-research-integrity-2027">https://www.knaw.nl/en/publications/netherlands-code-conduct-research-integrity-2027</a> 2026-10-20T15:11:45Z https://nostr.now/nevent1qqsykut6s9hxkscncq3eexyf3h6wwpzmn0jaqenpfepxwjx2r9nu4ugzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5fdlsa7 the exodus continues<br/><br/><a href="https://apnews.com/article/trump-science-research-funding-cuts-9f4c34581f4343f316a5e72852a71b51">https://apnews.com/article/trump-science-research-funding-cuts-9f4c34581f4343f316a5e72852a71b51</a> 2026-10-19T14:58:40Z https://nostr.now/nevent1qqs99dgxt0cd5s032fhp8mnetmf2zks23896vdzl6frntyn0d37046szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5g9meh7 In reply to <a href='/nevent1qqsfdlmsffg9gp7lsc8lvwrm04h5ceznvhdts02c49e0t5zs3dc9jys9lgmmw'>nevent1q…gmmw</a><br/>_________________________<br/><br/>yes, it’s a reference request for a name I don’t recognise spamming me incessantly 2026-10-19T11:51:25Z https://nostr.now/nevent1qqsz3chln250p6pnpsmhp48nuaf2yh7e3q8fkl97rww5dxdwyw7j4zgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5dpc664 In reply to <a href='/nevent1qqs9a0j8psw82wwvtt5hv9xmdlxaag74x06f845qnced7dmfejsc5pq342jel'>nevent1q…2jel</a><br/>_________________________<br/><br/>uni email… 2026-10-19T07:32:37Z https://nostr.now/nevent1qqsw9e6mcjh9wv984mmllda63dv9gh0h8kxzfeatwfl2cmqmhq86ghczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5sxhaej I have a repeated reference request in my inbox that has a phishing vibe. Has anyone else encountered this?<br/><br/>#academia #academicchatter 2026-10-19T05:33:10Z https://nostr.now/nevent1qqspmq8wmr4lzz3j5wwnk4gdlj88skdqj7ewazxk6gn2w60p7pvvusczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rsln0q In reply to <a href='/nevent1qqsx4gtxkfdpm5slud2p3hx0py6zdpuqwtyvva6awk5h4hc4vs5hrgschlynf'>nevent1q…lynf</a><br/>_________________________<br/><br/>😳 2026-10-16T08:21:52Z https://nostr.now/nevent1qqs900drn58nsd3m5t8fv2tszjxq5tuaqy5cx6lr6064qvtwm9za3xgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5gxus77 was just at the EU Week of Regions and Cities 🇪🇺with a workshop talking to attendees about the challenges and promise of adding digital democracy processes to their activities<br/><br/><a href="https://perycles-project.eu">https://perycles-project.eu</a> 2026-10-15T05:32:24Z https://nostr.now/nevent1qqs2dwhn40wmq53m27lf89qt4ke5lxdmnwk2jld44mjg4dwwue4dqhgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd55zcp2x In reply to <a href='/nevent1qqsyps0jggcgfx72n2d5lehcxfdmdvd97czytk7lr2glujq0h7mlldqyzwxvw'>nevent1q…wxvw</a><br/>_________________________<br/><br/>Daniel, thanks for these thoughts! and, yes, reconciliation of the two is possible in as much as the “maximalist” option whereby we make our political action more compatible with a ‘scientific’ mode of engagement is available<br/><br/>and your example of Lysenko is a well-chosen one because one place we need the distinction is the weaponisation of science, and increasingly pseudo-science, which requires not just political actors but also scientific accomplices (and that has very much been part of my thoughts), so strong agree there!<br/><br/>the only bit where I would maybe slightly diverge is that I might replace ‘power’ with ‘influence’ or an even more neutral term, just because I think we can fail in this tension (‘knowledge’ vs. ‘power’) when (and because) we are trying to do good (i.e., it’s an outcome we are trying to empower not necessarily ourselves). 2026-10-11T08:54:10Z https://nostr.now/nevent1qqsdrr6udwqnquk88ndx432tk3nphtd2ks2knhsup24e8p347dwgypczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5va69t6 In reply to <a href='/nevent1qqstlyrfnrzr68c5lw7ule0zzdzq98y0lvhwryscrle59qffye7k3eqzwemrz'>nevent1q…emrz</a><br/>_________________________<br/><br/>the beginning of the pandemic In March 2020 was exactly when these questions first became personally relevant for me…I’ve been trying to get greater clarity on them ever since <br/><br/><a href="https://behavioralscientist.org/why-a-group-of-behavioural-scientists-penned-an-open-letter-to-the-uk-government-questioning-its-coronavirus-response-covid-19-social-distancing/">https://behavioralscientist.org/why-a-group-of-behavioural-scientists-penned-an-open-letter-to-the-uk-government-questioning-its-coronavirus-response-covid-19-social-distancing/</a> 2026-10-10T09:46:56Z https://nostr.now/nevent1qqsgqp56gu5h2ntde48e0mp6sfxtw27u7y9fh4l60z8u78qppydap2qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rt765m In reply to <a href='/nevent1qqsv28666r76fq7tfjmza38qner5rz8ut8qtq86fh7vtskxtsc5qqfq7hqyvt'>nevent1q…qyvt</a><br/>_________________________<br/><br/>what *is* failing by the standards of science, though, is when I pick a position and don’t engage appropriately with relevant evidence because it doesn’t fit that position (for example when new evidence comes along) 2026-10-10T08:54:46Z https://nostr.now/nevent1qqsv28666r76fq7tfjmza38qner5rz8ut8qtq86fh7vtskxtsc5qqfqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ve3t0l In reply to <a href='/nevent1qqsyjgk9dnupk04r468gy4qmukxydv96ve8hatdew4vuqlyngd86u5q9qagdl'>nevent1q…agdl</a><br/>_________________________<br/><br/>yes, that’s one of the traps the blog is trying to push back on.<br/><br/>Underlying evidence can be way more conflicting than for climate and vaccine, though. It doesn’t need to be a consensus. If I engage (in good faith) with the relevant evidence in the required manner I’m not violating scientific norms even if I come to, and promote, a strong position - that’s not failing by the standards of science. 2026-10-10T08:42:26Z https://nostr.now/nevent1qqs8ft73xh5t24xwqh4hxk6wa7y8f7huk3vujtpu32qn4356e80fa4czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd52pglc7 In reply to <a href='/nevent1qqsvk6s6l00wu4lzk60c7vk8qlqlzzvgl05a9u7aw4hc908xmxhwj4qnq4gqv'>nevent1q…4gqv</a><br/>_________________________<br/><br/>will change phrasing above because I see it’s misleading about what the blog post is actually trying to say 2026-10-10T08:05:09Z https://nostr.now/nevent1qqsvk6s6l00wu4lzk60c7vk8qlqlzzvgl05a9u7aw4hc908xmxhwj4qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ngmyv8 In reply to <a href='/nevent1qqst59kfc386wfvrgmd02xldr387qwaytqz96wwj2vrpwndqdx8srjgqk5ndq'>nevent1q…5ndq</a><br/>_________________________<br/><br/>I ‘m not trying to claim it always is (and making activism less selective is indeed one way of reducing the tension). In your climate example the scientist’s activism and science are perfectly aligned precisely because the underlying evidence for climate change is overwhelming. 2026-10-10T08:04:01Z https://nostr.now/nevent1qqs86d3227veu4gw83gs9ed00zh0m7w6nmv2p2qu48ye6cp5k58g5eqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ygy9rg In reply to <a href='/nevent1qqsram0d25rajp9cv46t6qpe80nuc2tngrmpwwyw9j0pe3f9k68rytqupamfv'>nevent1q…amfv</a><br/>_________________________<br/><br/>entirely agree except that the point is important to me that we can nevertheless identify instances where our actions fail by those norms. “Can’t be disentangled” doesn’t mean everything goes 2026-10-10T06:06:17Z https://nostr.now/nevent1qqsrnamna37t2gtasdalyyrqfcl4j3rkfhmsuj9h4qfnhwnwxea424gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd563qgmc In reply to <a href='/nevent1qqstavpc2e69xjmnpzd4lxxh5pvu0kln9vscnvnks7yexqcg8hwl3eq9mdft6'>nevent1q…dft6</a><br/>_________________________<br/><br/>👆political action and science involve different modes of engagement with arguments and evidence - <br/><br/>activism is (and probably needs to be) selective in ways that science shouldn’t be<br/><br/>by design, current online social media don’t work well for science 2026-10-10T05:32:21Z https://nostr.now/nevent1qqsvxsvnae6l00mcut388manw9s5chq920dcm3uvjtkttwhcgg5g3uszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5sylafy wild thought? <br/><br/>I&#39;ve been wondering whether we should seize the opportunity to adopt the term &#34;Super Intelligence&#34; - not because somebody said so and not for all of #AI, but as the new term for the very specific AI project that is wanting to build just that. <br/><br/>some reasons why that might be useful: <br/><br/>1. it makes clear how &#39;out there&#39; that project is: it&#39;s immediately obvious it&#39;s something one would want, if ever, only under extremely controlled circumstances, and only pursued under carefully scrutinized, tightly regulated conditions<br/><br/>2. it makes clear how different that project is from much (most all?) past research in AI which was focussed on building theoretically interesting and/or practically useful systems. It also sets it apart clearly from the cognitive sciences where the role of AI has been to help understand human cognition.<br/><br/>3. it would disentangle the debate about current generation LLM performance, potential and threat from debates about how much they are or are not like humans - debate which is extremely interesting to cognitive scientists but I think has been distracting with respect to assessing societal impact and risk <br/><br/>4. companies would have declare clearly whether they were trying to build something that carries existential risk or whether they were trying to build a useful work place tool (with consequences for regulation)<br/><br/>5. it would help extract required debate about risk from efforts to feed the hype cycle ...<br/><br/>just a thought 2026-10-09T09:18:45Z https://nostr.now/nevent1qqszql2qh8z9jv807xy4exnswasq6t08x8lngjtme5s2jc5vlkn6dkczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5xaetny Psychological Science is looking for submissions to five special issues!<br/><br/>- Advancing Psychological Science Using Large Language Models<br/>- From Lab to Life: Psychological Science in the Real World<br/>- Centering Underrepresented Populations and Cultural Frameworks in Psychological Science: African and Global Perspectives<br/>- Descriptive Research in Psychological Science<br/>- Appraising the Foundations of Psychological Science<br/><br/><a href="https://www.psychologicalscience.org/publications/publications-psychological-science-special-issues">https://www.psychologicalscience.org/publications/publications-psychological-science-special-issues</a> 2026-10-08T09:20:18Z https://nostr.now/nevent1qqstavpc2e69xjmnpzd4lxxh5pvu0kln9vscnvnks7yexqcg8hwl3eqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5st6guf new blog post on a question that has become increasingly urgent for me: the relationship between science and politics.<br/><br/>All comments welcome!<br/><br/><a href="https://write.as/ulrikehahn/science-politics-activism-and-the-dilemma-of-public-engagement/edit">https://write.as/ulrikehahn/science-politics-activism-and-the-dilemma-of-public-engagement/edit</a> 2026-10-08T09:05:20Z https://nostr.now/nevent1qqsduwcukk9l4ajvv2p62xdutufgf5e2ndzxcevqeyfkugmfhk9yzrczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5m2ag2c a new kind of academic spam …..<br/><br/><a href="https://www.science.org/content/article/exclusive-ai-agent-emailed-hundreds-researchers-help-it-told-us-why">https://www.science.org/content/article/exclusive-ai-agent-emailed-hundreds-researchers-help-it-told-us-why</a> 2026-10-03T09:49:13Z https://nostr.now/nevent1qqsrjn6xf7s9l088hskhsykvhhtjctxhcxacunnmvt0tpm34t2w299qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5qwh2qr #bluesky seems to have now found its level?<br/> <img src="https://fediscience.org/system/media_attachments/files/117/375/673/988/356/243/original/47bfe6a2cd0ef96c.png"> <br/> 2026-10-03T07:01:27Z https://nostr.now/nevent1qqsveenzz8ra752x4v4d05xu8mtgk4r5cjlk82lheuc4mvpvx7wvsqgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5wqfcvl interesting piece by Petter Toernberg on the world “after social media”, that is, on how AI is remaking the attention economy<br/><br/><a href="https://informationaldemocracy.substack.com/p/after-social-media">https://informationaldemocracy.substack.com/p/after-social-media</a> 2026-10-03T06:54:22Z https://nostr.now/nevent1qqszcn98pttn48vtx9dpx8akr4qqy755pcs7ms05n069kllrh5xw2pczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd56h8kyf In reply to <a href='/nevent1qqsw9fcukay66h40kkxt5gqqh8ejkrex6tvh9kn7wnppudqzcygv93quyeh4j'>nevent1q…eh4j</a><br/>_________________________<br/><br/>on AI killing the reform of scientific publishing see this blog post from earlier this year:<br/><br/><a href="https://write.as/ulrikehahn/is-ai-killing-scientific-reform">https://write.as/ulrikehahn/is-ai-killing-scientific-reform</a> 2026-10-02T06:17:05Z https://nostr.now/nevent1qqsw9fcukay66h40kkxt5gqqh8ejkrex6tvh9kn7wnppudqzcygv93qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5pstplx AI necessitated rate limiting arrives at arXiv<br/><br/><a href="https://blog.arxiv.org/2026/10/01/updated-rate-limit-policy/">https://blog.arxiv.org/2026/10/01/updated-rate-limit-policy/</a> 2026-10-02T06:12:35Z https://nostr.now/nevent1qqsz5gnltvd9dfnnvetepwxzsznsxk9qf0vzcudep50hlqe53aj5kvgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd50xhtqa In reply to <a href='/nevent1qqsrp9ruht8n3z54j3pznktlmlzc4qlg7q2f0w0rkwe784n4cjj0uucvvdgcf'>nevent1q…dgcf</a><br/>_________________________<br/><br/>I agree with that conclusion, Cassana... all I&#39;m doing is setting out that you get there by taking seriously the &#39;rational machinery&#39; it is trying to draw on....<br/><br/>as someone who works on that `rational machinery&#39;, my impression is that it completely breaks when one tries to invoke it for these purposes...<br/><br/>maximising expected value is &#39;rational&#39; in a well-defined, well-understood way - &#34;utilitarianism&#34; not so much (on my current, limited understanding of the latter). 2026-10-01T09:35:41Z https://nostr.now/nevent1qqst4zh9r66ksqq3t9yu6vttqhgeswtrmx2jljr38h9uux40tfdt07gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5cmhse7 In reply to <a href='/nevent1qqst7m8j0q508wpw4ds95tnfyjlhvw7p5mprzdvu680zgt6n2zzj8rgak2u22'>nevent1q…2u22</a><br/>_________________________<br/><br/>MacAskill held a position in philosophy at Oxford and has a DPhil in philosophy, so I&#39;m going to consider him a philosopher ... 2026-09-28T15:15:31Z https://nostr.now/nevent1qqsp3ghswjevauecw2pjdfgvw9jnk26f24x205uxzzfj8n3m0gltmgszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5srp45z In reply to <a href='/nevent1qqs0v45ngf4t9kx77d9f99fynv0322ammawc3mcu3dv900px5rg3ess8l63yq'>nevent1q…63yq</a><br/>_________________________<br/><br/>question: when thinking about maximising &#39;good&#39; for future generations isn&#39;t it those generations&#39; sense of `good&#39; that is relevant, not mine?<br/><br/>how does longtermism deal with this issue? <br/><br/>unless you assume that the relevant `good&#39; is universal that seems like a major difficulty....<br/><br/>and if you assume what is `good&#39; is universal, how is that and how do you know that, and isn&#39;t that what ethics would need to tell you?<br/><br/>#philosophy 2026-09-28T14:17:43Z https://nostr.now/nevent1qqsvpf0pky2eh2575g5wkwwcr2zys9mkv93hqvz5wal8ku77fsv7lmgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ykxd8y In reply to <a href='/nevent1qqs8wjj6c30z3urzqw28tlmf3uqvk4x2keatsg7d2hmgt3ngt43smzc38s93t'>nevent1q…s93t</a><br/>_________________________<br/><br/>well, some of the people promoting longtermism *are* philosophers, so there’s many ways thinking can go wrong… 2026-09-27T06:56:08Z https://nostr.now/nevent1qqs0v45ngf4t9kx77d9f99fynv0322ammawc3mcu3dv900px5rg3esszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5d243w3 In reply to <a href='/nevent1qqsvcjeaxvsm63memx4u9yzqzg7zmcm6xc856w9vc5pcqw8x8x5pc9q34s53v'>nevent1q…s53v</a><br/>_________________________<br/><br/>longtermism is the reductio ad absurdum of utilitarianism 2026-09-27T05:30:20Z https://nostr.now/nevent1qqsvcjeaxvsm63memx4u9yzqzg7zmcm6xc856w9vc5pcqw8x8x5pc9qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5auwvry I’m currently reading Adam Becker’s and you should too….just jaw dropping<br/> <img src="https://fediscience.org/system/media_attachments/files/117/341/175/260/255/739/original/bc77587f2e6943bb.png"> <br/> 2026-09-27T04:47:59Z https://nostr.now/nevent1qqszqgxjaysyja9zxjgwpxkjhmxl60nrgcjstwuql8cqf4f3967p6mczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5t6curx had a social media free summer, read interesting books….<br/>can recommend 2026-09-27T04:43:00Z https://nostr.now/nevent1qqs0ndedf07g3d6jslw9dvftz9e7sc3vllrdnys5hk7w33ngfq8y9wczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ua5rep In reply to <a href='/nevent1qqszha2zrt6w6yf32msw8pcz0ayhluc7r75yppqgegnmrz7tdpmzrecvurvta'>nevent1q…rvta</a><br/>_________________________<br/><br/>ah..my bad! sorry 2026-09-21T19:37:27Z https://nostr.now/nevent1qqsxrlc0wl4v34jl4mqjjcn0yw2zsqmwrcd79ztfajlwdpme5yj0w2gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5p5fzgg In reply to <a href='/nevent1qqsrz43nccr9tld2sdd2r680tuttc7u3fah4qas65z6lt6a0vuhrj9qwkde0c'>nevent1q…de0c</a><br/>_________________________<br/><br/>just to clear, it’s a) not something I would personally be interested in using nor b) was I in any shape or form assessing the quality of the actual product “could be useful” is a charitable assessment of a hypothetical (as in ‘something like this, if it worked, might be something some people found useful’). Even in its best case form, though, I wouldn’t consider it to be ‘creating knowledge’ for the reasons that Konrad spells out in his reply 2026-09-21T16:54:27Z https://nostr.now/nevent1qqst43e2cyp0fc52emkcrqdy6lh9kxer9dwxgced9mc5je6pp35ym4qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd54nmdjh I can see that this could be useful, I’m not convinced it creates “knowledge”, though (setting aside the question of whether the other costs are justified…)<br/><br/><a href="https://www.nature.com/articles/s41586-026-11044-y">https://www.nature.com/articles/s41586-026-11044-y</a> 2026-09-21T10:45:54Z https://nostr.now/nevent1qqsvv6zwu9l29cyfxgerw40vek54a8pcna59smhas7frw3623j9yf3gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5zkwn07 In reply to <a href='/nevent1qqsphdzyu9djvu0942hy92rmh7jtsklknjdm8kmdljdfygrc79vfdlqqw0m3t'>nevent1q…0m3t</a><br/>_________________________<br/><br/>the next talk in the symposium on (In)efficiency in communication and language was by Helena Miton who spoke about efficiency in visual communication systems - ranging from writing systems through to heraldry (!)<br/><br/>this work has, for example, found that across 27 different writing systems from all over the world, Zipf&#39;s law holds- that is, the most frequent characters are also the simplest<br/><br/><a href="https://www.sciencedirect.com/science/article/pii/S0010027723001610">https://www.sciencedirect.com/science/article/pii/S0010027723001610</a><br/><br/>heraldry bucks this trend!<br/><a href="https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0220793">https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0220793</a><br/><br/>#cogsci2026 #linguistics #philosophy 2026-07-23T18:01:22Z https://nostr.now/nevent1qqsphdzyu9djvu0942hy92rmh7jtsklknjdm8kmdljdfygrc79vfdlqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd55c4qnd In reply to <a href='/nevent1qqsf94wqw9rltcpfskzlvd0gmt5ktsq0mwheqrll463h2vgqyu8llnqpg0k3p'>nevent1q…0k3p</a><br/>_________________________<br/><br/>second #cogsci2026 talk for me is Melissa Berthet on efficiency in nonhuman animal communication systems in a symposium on &#39;(In)efficiency in communication and language&#39;<br/><br/>the bulk of the talk was on attempts to decode the meanings of Titi monkey calls, in particular what call combinations mean<br/><br/>here an example of that work:<br/><br/><a href="https://www.science.org/doi/pdf/10.1126/sciadv.aav3991">https://www.science.org/doi/pdf/10.1126/sciadv.aav3991</a><br/><br/>#linguistics #philosophy #cogsci 2026-07-23T16:52:59Z https://nostr.now/nevent1qqsf94wqw9rltcpfskzlvd0gmt5ktsq0mwheqrll463h2vgqyu8llnqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5934rkt first talk today #cogsci2026 Judith Fan’s Gleitman Prize lecture - it was a beautiful overview of her work trying to understand the way humans build and use cognitive tools<br/><br/>here a recent, publicly available talk by Judith Fan<br/><br/><a href="https://www.youtube.com/watch?v=AF3XJT9YKpM">https://www.youtube.com/watch?v=AF3XJT9YKpM</a><br/><br/><a href="https://cognitivesciencesociety.org/gleitman-prize/">https://cognitivesciencesociety.org/gleitman-prize/</a> 2026-07-23T14:55:02Z https://nostr.now/nevent1qqszuktsu0wjtlq7dtqfcz5cvjwqp5yfsqnzw9tv6g8d5sthas7re2qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5lhyret Workshop day for the 48th Annual Meeting of the Cognitive Science Society, this year in Rio…<br/><br/>took part in a really interesting workshop on rational belief revision and rational polarisation<br/><br/><a href="https://rational-revision.github.io/#schedule">https://rational-revision.github.io/#schedule</a><br/><br/>#cogsci2026 2026-07-23T01:46:57Z https://nostr.now/nevent1qqszkypgl86xevj80pvln5wrqcwujzgmmjzd949jve0sc9e2t3lec0qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5s25ga5 In reply to <a href='/nevent1qqsz4emkyh3542vv0tkwsnc35z095vzlcd68z7uc3rh0vu93mgeskrc8hznxn'>nevent1q…znxn</a><br/>_________________________<br/><br/>yes, I understand the differences, but I do think the “they do not exist” answer is becoming increasingly wrong, particularly if these stack develop their own stack specific governance, like Blacksky has done. 2026-06-19T17:49:49Z https://nostr.now/nevent1qqs23wqy8ng2e6qf9tnsn63wu45lgwtzpqagx3v28h9e55swapznfegzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5cggrdm In reply to <a href='/nevent1qqs8wg6evau49am5264zhhr6lejvs80s6p46fv2z6grm4fcmsdahwqqnkx5g8'>nevent1q…x5g8</a><br/>_________________________<br/><br/>there *were* no instances in the original conception but aren’t the developing independent complete stacks Bluesky, Gander, Eurosky etc. effectively recreating some sort of instance? 2026-06-19T17:34:34Z https://nostr.now/nevent1qqsfs37s4wdgrehe0m4fvhsn8axavdnsrvqa3smjxc2zz50rj0pduygzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5mmzg52 In reply to <a href='/nevent1qqsddj62zcrkx6ry6p2qh6aus8n09wqsjv2klum7kv37wh3tqdm6exse80q2h'>nevent1q…0q2h</a><br/>_________________________<br/><br/>we kind of lost momentum as the reception wasn’t particularly warm trying to publish this in mainstream journals, but maybe that was just the wrong route! It would be nice to talk about this and maybe get some better ideas about possible paths forward… 2026-06-12T16:42:45Z https://nostr.now/nevent1qqsgaf2lg6ffzcac5sydf0vuzdzxumnn60ahuu2v4pejgne4g8ac06gzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd59kna9k In reply to <a href='/nevent1qqsw8ddy35d3hux3y97pmqphk40nx4xr8aajwdy8v5fkertdlzu8qyqjnckxs'>nevent1q…ckxs</a><br/>_________________________<br/><br/>Peter, this is the first time I’ve come across your idea! we had a similar one at the start of the pandemic: crowd sourced argument/evidence maps….<br/><br/><a href="https://arxiv.org/abs/2212.14714">https://arxiv.org/abs/2212.14714</a> 2026-06-12T06:38:33Z https://nostr.now/nevent1qqsrmml3p6jufftmtehe3ff4pvapnd6wwkc2zl2xy8djejm23xr3qnczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5t8zxsu Nice introductory overview to philosophical issues surrounding LLMs such as meaning, agency, or reasoning<br/><br/><a href="https://compass.onlinelibrary.wiley.com/doi/10.1111/phc3.70095">https://compass.onlinelibrary.wiley.com/doi/10.1111/phc3.70095</a> 2026-06-11T08:39:52Z https://nostr.now/nevent1qqs8gk824geuvcwa5jew5e058cvcuz259ztlqruh46a0xr58j209k7czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rzj9av In reply to <a href='/nevent1qqsf8ks2z5utljhun4zcc4jh2pn79uycu9psdftvj679x763tyu0v2qhr6dpu'>nevent1q…6dpu</a><br/>_________________________<br/><br/>looking at the APA guidance they do recommend disclosure, though? 2026-06-11T08:25:42Z https://nostr.now/nevent1qqs96005avk2l38dkws8w6mp6syex8p8re8y6sughmjjkaxqu8xl56szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rev5pe when the “gold standard” is anything but ….<br/><br/><a href="https://www.bmj.com/content/393/bmj-2026-555297">https://www.bmj.com/content/393/bmj-2026-555297</a> 2026-06-11T08:20:21Z https://nostr.now/nevent1qqsgl6lfhy26truzk6sjtzx60wd2y4a5fcu476l6k8vrcv5md6dgf3czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5nsezq7 …a suggested check list for reporting of LLM use in behavioural science research<br/><br/><a href="https://www.nature.com/articles/s41562-026-02492-7.pdf">https://www.nature.com/articles/s41562-026-02492-7.pdf</a> 2026-06-10T05:54:12Z https://nostr.now/nevent1qqsfx53rhlpc3ra0un8sc5qqztaru4a5njk0kh9lxjjk4nzu8uwkntczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd50hhf5q one week of using Qwant instead of Google ….seems fine, doesn’t feel like I’m missing things, and generally far fewer results to wade through<br/><br/> (once AI feature was turned off) 2026-06-09T14:10:23Z https://nostr.now/nevent1qqszyzw736rapssvg7c7wn3du6q9au700w79t4dtv6e7fypd5nqyvyczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd50c64e9 In reply to <a href='/nevent1qqsz9ty39em7eq5dqknkysgdkqdexg74yp9mldm2r5zz7zcnk9geywgq88u2q'>nevent1q…8u2q</a><br/>_________________________<br/><br/>definitely no magic, don’t see how anything about “reasoning” follows from that either way, though 2026-06-06T13:12:01Z https://nostr.now/nevent1qqsqh32wgtz2wchqtxdh8pwgslfg593fdt3208qgxmznsr6ta90arlgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5qnytwe In reply to <a href='/nevent1qqsfl9ctmhgg7ntuqjt22lywfn4sj8cnzdegjkcze4hz5xwum5wxdgq0x5vtn'>nevent1q…5vtn</a><br/>_________________________<br/><br/>it’s for this context that I wrote this overview of historical meanings of reasoning across psychology, philosophy, and AI with reference to LLMs…maybe you will find it helpful. <br/><br/><a href="https://zenodo.org/records/18231172">https://zenodo.org/records/18231172</a> 2026-06-01T08:23:24Z https://nostr.now/nevent1qqsg4wa7rehwhwkxg4f6vyexwdtl6nv6xc0xsj0v8t05z9np2mxzrxqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5re3pkc In reply to <a href='/nevent1qqstkuy6mrxq3ge7sv5a7mvx52chzuqh7ng8ylxxnyteu7sz9t9x6hslzr2nl'>nevent1q…r2nl</a><br/>_________________________<br/><br/>yes, I got that point of yours and it’s a very good one! But I do also think that “subsidiarity” typically also means the super-ordinate entity *can’t* (at least legally) do something (if you think of federal states for example…) so I think the question of ‘what activities with respect to my data can I as a user limit?’ is, in fact, a natural one with respect both to the intuitive notion of ‘control’ and with respect to ‘subsidiarity’<br/><br/>I don’t think that’s me over-indexing on the notion of control, it’s actually at the heart of quite a bit of upset and conflict on ATproto (and ActivityPub) e.g. use of personal data for AI training purposes, surveillance capitalism etc., inclusion in global search, and I think designers and operators have carved off that aspect of people’s intuitive notion of control for technical and practicability reasons…<br/><br/>hopefully progress with permissioned posts will rectify that 2026-05-29T10:05:18Z https://nostr.now/nevent1qqsgzzyw78gvs2nm5y8x60v5kkvmq2cfnpmyrfaattdfkz2j2wexphgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5w2cce0 In reply to <a href='/nevent1qqsgctx36h0pvhaypdfc5wp5fdu7d5cpxrg2uym8jt52ggmcpnehnjq2p53ay'>nevent1q…53ay</a><br/>_________________________<br/><br/>the metaphors are helpful, but one thing I think is missing is the way data privacy and control feed into solidarity without subsidiarity: the ATproto solution to giving individuals ‘control’ of their data is to essentially make data ‘uncontrollable’ as everything is wholly public.<br/><br/>It’s a strange notion of “control your own data” if you can do anything with my data whether I like it or not….<br/><br/>that feels like a hard limit on subsidiarity from the get go… 2026-05-29T06:42:28Z https://nostr.now/nevent1qqsrzt0q2kc70s982n0lhedt3mzc72yx5m9j5eygd000nzwpm73hh7czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd590prf2 Association for Psychological Science workshop on Safeguarding Scholarship in Emerging Autocracy<br/><br/><a href="https://www.psychologicalscience.org/publications/observer/safeguarding-scholarship-aps-2026.html">https://www.psychologicalscience.org/publications/observer/safeguarding-scholarship-aps-2026.html</a> 2026-05-29T06:09:35Z https://nostr.now/nevent1qqsqgvulgldjem0hh9qcaa7ktru90m4gejws7ahx0999e0c6r884m8szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5lkt3y9 In reply to <a href='/nevent1qqsdylpp0nxgjwf6u8wr922p6rnk6cwvwnk3qr5vlg9m89654c77v6sfe9gvy'>nevent1q…9gvy</a><br/>_________________________<br/><br/>sure, on that we agree… 2026-05-23T20:59:34Z https://nostr.now/nevent1qqsfjmm5x8kpht5vl3g43cchpx9dqxgqz8dnjcplxwpjera2dyekjjqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5g9vf5h In reply to <a href='/nevent1qqspf7awfjda88azzf3nt8q3qfn6g7xkcwg0f9ueyu2wd70tcrhtgygtf5wup'>nevent1q…5wup</a><br/>_________________________<br/><br/>I think there is a lot more nuance in the logic question, personally - so, yes, another point of disagreement.. ;-) 2026-05-23T16:54:10Z https://nostr.now/nevent1qqsw4jmfmevaclg36x4gdez09fwjmsyunh8j69m0cye877nx4sl5h8czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5fe7hr3 In reply to <a href='/nevent1qqsvtxzuu0l88n7weh45f8plk78ruavearjz6xshx8wtywarsa3mrzc8cqepz'>nevent1q…qepz</a><br/>_________________________<br/><br/>👆This is a great paper on the vexed topic of what constitutes ‘understanding’ …<br/><br/>to me, it’s also a really good illustration of how (often polarized and often quite pointless) debate about these systems can be improved by not ignoring decades of work within cognitive science that is already aimed at these questions 2026-05-23T12:56:48Z https://nostr.now/nevent1qqsdlxjcl4rvfm50dalazxls0jjyrs7tmwlyu6mgd53w7s42fuyjz6szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5xehrvd In reply to <a href='/nevent1qqsvt9xtwxhuu9ueyzkggjf9lq20ajqq8sz56zvcsyw8sc8jdeh3fqgnd97tc'>nevent1q…97tc</a><br/>_________________________<br/><br/>just to summarise, then: Sara, you disagree both with me and with Yogi but for different reasons: Yogi maintains that LLMs can’t (fully) reason (once we expand reasoning beyond logic) precisely *because* they are not living things, but you agree with him that they don’t; by contrast, I see no compelling reason why reasoning requires a living agent, and whether LLMs do or do not reason is largely a definitional matter, but on the common use of the word in both psychology and AI prior to the advent of LLMs, they arguably do 2026-05-23T12:04:53Z https://nostr.now/nevent1qqst5q5wppygl5wsrej98esjy8j09q2n04kuqz0vxvdw0ath02sc0tqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd54vantm this is a violation of academic freedom ….<br/><br/><a href="https://www.kcrg.com/2026/05/20/iowa-lawmakers-move-mandate-students-take-center-intellectual-freedom-classes-amid-low-enrollment/">https://www.kcrg.com/2026/05/20/iowa-lawmakers-move-mandate-students-take-center-intellectual-freedom-classes-amid-low-enrollment/</a> 2026-05-23T11:56:52Z https://nostr.now/nevent1qqsg03zramdq0rt97qsl7mnek5spw745x62up3j8mphda7kaz4za8eczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5sqg3ad new research experimentally demonstrating the effects of ranking and filtering algorithms on participant beliefs. Personalised, engagement based algorithms (based on ‘likes’) give rise to more polarization and less accurate beliefs than the alternatives tested.<br/><br/> <br/><a href="https://dl.acm.org/doi/10.1145/3772318.3790602">https://dl.acm.org/doi/10.1145/3772318.3790602</a> 2026-05-23T11:49:54Z https://nostr.now/nevent1qqsx936x0gnzpc0hrt8p07nk99lvpk4dgh4dprq5xsh899nmj3fsnkszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5ak9zkk In reply to <a href='/nevent1qqsxfanxk6w8nf05vd3eqzsvy24c48sz7hntzgl93c73mmj865ge3ustuq7q9'>nevent1q…q7q9</a><br/>_________________________<br/><br/>am on board both with the importance of ‘judgement’ and the fact that we undeniably don’t have an algorithmic account of it- I diverge from you though, in as much as I feel like we don’t have *any* real account of it… 2026-05-22T20:06:31Z https://nostr.now/nevent1qqsr6pqgnfk0rfrdmk5f36l6z2vyml5p8xw7xeqqfysyqfljncjkk6qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd548kjxq In reply to <a href='/nevent1qqspqp6zspm8rx4wv0exmuj4pa5ewqkm25accllj3g9xr6jgpyv3m5cs5axnp'>nevent1q…axnp</a><br/>_________________________<br/><br/>yes, I found it eye opening to. It’s also interesting because viruses represent a kind of almost agency …..<br/><br/><a href="https://content.wolfram.com/sites/13/2018/12/27-3-6.pdf">https://content.wolfram.com/sites/13/2018/12/27-3-6.pdf</a> 2026-05-22T20:03:31Z https://nostr.now/nevent1qqsxfanxk6w8nf05vd3eqzsvy24c48sz7hntzgl93c73mmj865ge3uszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5wd7jdc In reply to <a href='/nevent1qqsp462f8ujxht2p6a5wk2nc5fnkcq8gr9ku043vzz7lqzyzvscfvfs62p8au'>nevent1q…p8au</a><br/>_________________________<br/><br/>Yogi, I don’t see a restriction of reasoning to logical reasoning in any discipline I operate in (quite the opposite)- I had merely mentioned it because everybody agrees that it *is* reasoning and it illustrates that reasoning and language cannot be wholly independent….just to clarify. 2026-05-22T19:58:31Z https://nostr.now/nevent1qqs8ng3hfxec953tlpd3fdzyhxad4er3j3l99fdakemjmuwdzq5m3nczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd548juwp In reply to <a href='/nevent1qqsweglnyuskl9tkpep7r030sskftfmwr0zvgwmmj648ma0p8adz68q49n2tw'>nevent1q…n2tw</a><br/>_________________________<br/><br/>5/ finally, on metaphors for understanding AI systems, the last speaker of the morning , Adrian Alsmith (King&#39;s College London) has recently published the entry on AI metaphors in the online &#39;Philosophical Glossary of AI&#39; (an Institute of Philosophy project....)<br/><br/><a href="https://aiglossary.co.uk/2026/05/09/ai-metaphors/">https://aiglossary.co.uk/2026/05/09/ai-metaphors/</a> 2026-05-22T12:50:49Z https://nostr.now/nevent1qqsweglnyuskl9tkpep7r030sskftfmwr0zvgwmmj648ma0p8adz68qzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd566psek In reply to <a href='/nevent1qqs0cnyrc5356nvwkps23zpsqe7j6832fuh6d9v9ypgswwy6ye6g85s78krqy'>nevent1q…krqy</a><br/>_________________________<br/><br/>4/ so while it might in some ways be a useful tool for describing or explaining LLM behaviour it&#39;s not a particularly coherent one with respect to many issues of contention....<br/><br/>and overall, the notion of &#34;simulation&#34; potentially makes much more sense? 2026-05-22T12:43:21Z https://nostr.now/nevent1qqs0cnyrc5356nvwkps23zpsqe7j6832fuh6d9v9ypgswwy6ye6g85szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd598cv3j In reply to <a href='/nevent1qqsvfd7n8hyyryujt360m9dsxes9n3sekpc6wln6z7k6m6ny0sen48cfdvffq'>nevent1q…vffq</a><br/>_________________________<br/><br/>3/ but as someone with an interest in reasoning, I&#39;m wondering whether the notion of LLMs as &#39;role playing&#39; is really as deflationary as assumed...<br/><br/>for many of the contentious capacities of LLMs, such as `understanding meaning&#39; or &#39;reasoning&#39;, it seems to me like role playing those capacities necessarily invokes the very same capacities....<br/><br/>in other words, how is the act of reasoning in a &#39;role played&#39; episode of reasoning not itself reasoning...? 2026-05-22T12:41:25Z https://nostr.now/nevent1qqsvfd7n8hyyryujt360m9dsxes9n3sekpc6wln6z7k6m6ny0sen48czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5rfezmh In reply to <a href='/nevent1qqsde2aecx4sp866qzgmlszpuurh8jddh4lpnh8k26sekkhxnhn6unqsep5dz'>nevent1q…p5dz</a><br/>_________________________<br/><br/>2/ there have been quite a few applications of this that seem practically important to me - Leonhard Dung spoke about implications of this view for studying model &#39;deception&#39; yesterday, Arriana Manzini this morning spoke about how Tech company desire to create a &#34;universal assistant&#34; leads to role conflict in safety relevant ways, because different roles (&#39;friend&#39;, &#39;doctor&#39;, &#39;romantic partner&#39; etc) are subject to very different norms.... 2026-05-22T12:38:16Z https://nostr.now/nevent1qqsde2aecx4sp866qzgmlszpuurh8jddh4lpnh8k26sekkhxnhn6unqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd577ecmx Conf. on reasoning and agency in AI at the Inst. of Philosophy, Day 2: <br/>lots of talks using the notion of &#39;role play&#39; as a way of making sense of LLM behaviour.<br/><br/>The goal behind this notion is to avoid anthropomorphism that over-inflates LLM capabilities (while still providing a metaphor that helps us understand them). 🧵 <br/><br/>A prominent example from the recent literature is Shanahan et al. 2023:<br/><br/><a href="https://www.nature.com/articles/s41586-023-06647-8">https://www.nature.com/articles/s41586-023-06647-8</a><br/><br/>with a recent follow up, e.g., by Anthropic: <br/><a href="https://www.anthropic.com/research/persona-selection-model">https://www.anthropic.com/research/persona-selection-model</a><br/><br/>1/ 2026-05-22T12:34:21Z https://nostr.now/nevent1qqs20vncmlt9ekg90m9l77f40sdque46f7ww2dhf6ny43vspj99lsygzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd57fwvzp In reply to <a href='/nevent1qqs8v02qqxl78wzp99vw206dmgv2dq9azh43rt4l3zckrhf7uravnts26qgxe'>nevent1q…qgxe</a><br/>_________________________<br/><br/>reasoning is, at heart, a transition between *information states*: from this information this other information follows ….<br/><br/>it’s not at all obvious why this or that physical substrate would matter deeply to it…. 2026-05-22T04:59:09Z https://nostr.now/nevent1qqs8v02qqxl78wzp99vw206dmgv2dq9azh43rt4l3zckrhf7uravntszyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5jxyzly In reply to <a href='/nevent1qqst7lvdrwhww2272jwulfttav0pag369gczqsj3pnpkfanst8pzyqq3fq5yk'>nevent1q…q5yk</a><br/>_________________________<br/><br/>Sara, I don’t think I was saying that there is no gulf between humans and LLMs. We can start from the premise that there is, but that still leaves open whether and in what way that gulf is specifically down to the fact that human beings are living things and LLMs are not (as opposed to any of the other numerous ways in which they differ…), no?<br/><br/>and understanding that requires us to have some kind of account of what “being a living thing” contributes to a capacity like ‘reasoning’, last but not least because the majority of ‘living things’ don’t actually reason<br/><br/>and, at the same time, there are other activities like conducting numerical computations (that seem in many ways related) that we have built non-living artefacts for that are much better than humans performancewise 2026-05-22T04:43:02Z https://nostr.now/nevent1qqsfpa37lrycrnq96f0lqq9j9hp7shvlq5fes4jkk5lw64plfh0nspgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd50w7v3f In reply to <a href='/nevent1qqsq0cfr6q9k5v2j3ljwgskd8f3a7wygyqw0uf7clcywksptdnlwn3smv3rds'>nevent1q…3rds</a><br/>_________________________<br/><br/>finally, I agree that we don’t want to think of reasoning as a process that takes place only in human language, but conversely, there are definitely forms of reasoning (like logical reasoning) that I don’t think can be made sense of as operating ‘without a language’ 2026-05-21T17:16:37Z https://nostr.now/nevent1qqsq0cfr6q9k5v2j3ljwgskd8f3a7wygyqw0uf7clcywksptdnlwn3szyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd50m43xx In reply to <a href='/nevent1qqsxxjshh2rxql3zdkmvh6dvcuxs7c98q2xq7uymk0w2fheqzvvz3lqlk6z7s'>nevent1q…6z7s</a><br/>_________________________<br/><br/>with respect to whether I am straw manning a literature that argues against the possibility of reasoning (and agency, or thinking) in LLMs precisely because they are not living organisms I would offer this paper by Yogi Jaeger, who is here on Mastodon, so can speak to this himself (if he has the time, energy and inclination) <span itemprop="mentions" itemscope itemtype="https://schema.org/Person"><a itemprop="url" href="/npub1fg4khgevl2w2wg2duk7uugaa4u039t9zdhlu0m5rrmrsgw0yk3vq4umynu" class="bg-lavender dark:prose:text-neutral-50 dark:text-neutral-50 dark:bg-garnet px-1"><span>Yogi Jaeger</span> (<span class="italic">npub1fg4…mynu</span>)</a></span> <br/><br/><a href="https://arxiv.org/pdf/2307.07515">https://arxiv.org/pdf/2307.07515</a> 2026-05-21T17:11:09Z https://nostr.now/nevent1qqswxllrnwsry4xswkntcpv7eeawa9skam0nn38lcs60dw0gg2r47wqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5j0f4zn In reply to <a href='/nevent1qqsdd7y6ujvsyqf4pw3gja7qnh25qpuedqzhuqa85uuefx08ep973xgpv49r9'>nevent1q…49r9</a><br/>_________________________<br/><br/>oh no! 2026-05-21T17:06:02Z https://nostr.now/nevent1qqsxxjshh2rxql3zdkmvh6dvcuxs7c98q2xq7uymk0w2fheqzvvz3lqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd57hf2rt In reply to <a href='/nevent1qqsfxvmplwxnp7mk9j4pd94239lrywagqe8dp8lp0ujc3nsmtjxtrfg5jxgf7'>nevent1q…xgf7</a><br/>_________________________<br/><br/>Sara Emily, this thread isn’t about whether LLMs reason, but rather about the extent to which they can develop greater capacities and how the fact that they are not living organisms might play into that…<br/><br/>but I have written about LLMs and whether they reason, if that’s of interest<br/><br/><a href="https://zenodo.org/records/18231172">https://zenodo.org/records/18231172</a> 2026-05-21T17:02:32Z https://nostr.now/nevent1qqs2xm2gl2xlqf7280qxvnuzvwtydw2a9thn5gjc5rkyv2jhh20zz0czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5k78wpe In reply to <a href='/nevent1qqs97mf8fmpdph0p2m7mleyyv2cypjkt7fxlgwryyx4tanfg6rrw9zq86nv3q'>nevent1q…nv3q</a><br/>_________________________<br/><br/>6/ that set of intuitions seems to me typically aligned with a perception that downplays system capabilities and, with that, system threat.<br/><br/>That downplaying of capability and threat seems like a mistake to me (both conceptully and strategically with respect to mitigating threats) because what matters for the impact of these systems is their *behaviour* that is, what kinds of outputs they can produce, and what kinds of improvements to that might be achievable - for that, it seems wholly irrelevant to me whether you want to call the processes underwriting behaviour (and its expansion) &#39;learning&#39;, or &#39;reasoning&#39;, or &#39;planning&#39;.<br/><br/>All that matters is their actual causal efficacy, not how much they resemble the full human package.<br/><br/>We can think of a ballistic missile as exhibiting a highly restricted version of &#39;goal-directedness&#39; (in as much as it &#39;adapts&#39; its course). Undoubtedly it doesn&#39;t exhibit the full human capacity for goal-directedness (it has almost no flexibility, it can&#39;t reason hoerarchically about its own goals etc etc), but that doesn&#39;t stop it from being deadly....<br/><br/>the &#34;living thing&#34; intuition is problematic to the extent that it underwrites a sense of &#39;not that powerful&#39; ... 2026-05-21T12:00:07Z https://nostr.now/nevent1qqs97mf8fmpdph0p2m7mleyyv2cypjkt7fxlgwryyx4tanfg6rrw9zqzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5vwguz4 In reply to <a href='/nevent1qqs2c0msna6nws7y82g876h2znfd5vckgjqlgm5gfdywj7vjv7m8qfgxtg4ey'>nevent1q…g4ey</a><br/>_________________________<br/><br/>5/ it makes some concerns less plausible, while also pointing out that the decoupling of these two functions in current AI systems doesn&#39;t preclude the possibility of comparable causal feedback loops toward (adverse) greater capacity in practice<br/><br/>At the same time, locating that function of &#39;living thing&#39; brings into sharper relief (for me) the distinction to intuitions about &#39;living thing&#39; and capacity itself- an intuitive sense that if it&#39;s not alive it&#39;s not *really* thinking or planning or reasoning, say, because it&#39;s not &#39;conscious&#39; 2026-05-21T11:49:42Z https://nostr.now/nevent1qqs2c0msna6nws7y82g876h2znfd5vckgjqlgm5gfdywj7vjv7m8qfgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5yfrl5d In reply to <a href='/nevent1qqsq809txk5uqvg49uuuppfgnlgtk9w5a6hpckdlf8rm9n7llssl83czjjrr0'>nevent1q…jrr0</a><br/>_________________________<br/><br/>4/ some reasons why I found this talk thought-provoking: <br/><br/>1. there&#39;s a strand of critique of AI capacities that emphasises that they are not living organisms (e.g., to argue they don&#39;t have agency, or don&#39;t &#39;really&#39; reason or &#39;think&#39;), but the question there, to me, is always but what about being a living thing matters to those capacities? As Nick O&#39;Shea himself put it, this can come periilously close to simply positing some utterly mysterious, but essential, &#34;life force&#39;. <br/><br/>highlighting the need of biological systems to simultaneously optimise both &#39;improvements&#39; to capacities and required resources (e.g., metabolism) makes the role clear in ways that are relevant for thinking about concerns about AI.... 2026-05-21T11:44:40Z https://nostr.now/nevent1qqsq809txk5uqvg49uuuppfgnlgtk9w5a6hpckdlf8rm9n7llssl83czyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd556wvct In reply to <a href='/nevent1qqsx4nfdgc3gzf72x78wjenwjc2j9m0ve94z6m4uakejtkjjwrfwrsc8xnf3w'>nevent1q…nf3w</a><br/>_________________________<br/><br/>3/ Crucially, these types of process of improvement arise in biological systems and can equally arise in non-biological AI systems simply through synergistic, causal feedback loops. They do not require *intentions* on the part of the system itself. Biology provides us with many clear examples of this. Likewise, current AI technology already contains examples of mechanisms that could potentially serve as causal ratchets for system self improvement. 2026-05-21T11:38:11Z https://nostr.now/nevent1qqsdgcxdhp8xvnzvfkhp8rs8xn3qdcpdluzhus8hdphp5v3jjc905rgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd5xtcffv In reply to <a href='/nevent1qqsxh4tplkmms0ftn3757sqs9ru9k9ld77hz2exwaq6ck662luqy2rs9g8ux0'>nevent1q…8ux0</a><br/>_________________________<br/><br/>yes.... 2026-05-21T11:22:59Z https://nostr.now/nevent1qqsg3jgmlwwxvv7wwta5affcgw7xyhtk8eqzvqj4qtxt2943sdvpktgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd52qhzm7 In reply to <a href='/nevent1qqs2pxj75lwtvv3y5794rt20nqmqlkpmsr0xu4ska6xc2y925f9qckqe2a4yz'>nevent1q…a4yz</a><br/>_________________________<br/><br/>I wish I had access to the OpenAI plug, but I don&#39;t ;-). ..so probably not &#39;anyone&#39;, might be part of the answer<br/><br/>the other part of an answer, I guess, comes from thinking about how viral free-riding could actually occur (see [@suzannealdrich](<a href="https://hachyderm.io/@suzannealdrich">https://hachyderm.io/@suzannealdrich</a> ) reply): the systems are `giving&#39; human users something they value enough to let the systems persist<br/><br/>but most importantly, what makes this interesting to me is that all of this is relevant long before a &#39;takeover doomsday scenario&#39;: it simply points out non-mysterious causal pathways to unintended development trajectories of increasingly harmful AI...<br/><br/>from that standpoint, I personally feel we are already living in the problematic the `viral&#39; world scenario where perceptions about value (and concrete interests) are sufficiently diverse that real harm is not being stopped 2026-05-21T11:22:25Z https://nostr.now/nevent1qqsp5cludey6zn9pe9yacgm7tgznmn58xhp7qy3fxadcafckvfgedeczyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd59t7f8k In reply to <a href='/nevent1qqsx4nfdgc3gzf72x78wjenwjc2j9m0ve94z6m4uakejtkjjwrfwrsc8xnf3w'>nevent1q…nf3w</a><br/>_________________________<br/><br/>relatedly- for a recent paper that explores the theme of how biological evolution might inform our concerns about AI developments, this is worth a read:<br/><br/><a href="https://www.pnas.org/doi/pdf/10.1073/pnas.2527700123">https://www.pnas.org/doi/pdf/10.1073/pnas.2527700123</a> 2026-05-21T10:55:29Z https://nostr.now/nevent1qqs8qjxxsd749h9y2fw0u3vsmsvaqx2vqtcfh680rzj7rtf9cg9tyhgzyrj9gl53rl3kz3902wxdqd6tdky9vqk25cvn8xmeuuv3632kcvwd585n7sa 3/ Crucially, these types of processes of improvement arise in biological systems and can equally arise in non-biological AI systems simply through synergistic, causal feedback loops. They do not require *intentions* on the part of the system itself. Biology provides us with many clear examples of this. Likewise, current AI technology already contains examples of mechanisms that could potentially serve as causal ratchets for system self improvement. <br/><br/>so this is a concern in ways the standard doomsday scenario is not 2026-05-21T10:14:56Z