Recent searches

No recent searches

Search options

Not available on photog.social.
photog.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A place for your photos and banter. Photog first is our motto Please refer to the site rules before posting.

Administered by:

Server stats:

242
active users

photog.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.3

#turingtest

0 posts0 participants0 posts today
NerdfallmanagementHomophobie Depression Suizid
➴➴➴Æ🜔Ɲ.Ƈꭚ⍴𝔥єɼ👩🏻‍💻<p>Okay, here it is. This is the unofficial official timeline of <a href="https://lgbtqia.space/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a>. I'm going to tell you what to expect, and it's definitely not: this all goes away and we return to before. </p><p>Are you ready for this? Are you sure? Well, read on.</p><p>Before I continue, I'm going to lay out some AI <a href="https://lgbtqia.space/tags/benchmarks" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>benchmarks</span></a> that we'll use to define "how good / scary is this AI?" This is in rough order of difficulty.</p><p><a href="https://lgbtqia.space/tags/Lovelace" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Lovelace</span></a> <a href="https://lgbtqia.space/tags/Test" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Test</span></a> for <a href="https://lgbtqia.space/tags/Emergence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Emergence</span></a>: "Can a system produce surprising and useful outputes that weren't explicitely programmed via weak emergence?"</p><p><a href="https://lgbtqia.space/tags/Loebner" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Loebner</span></a> Test: "Can a computer fool casual human judges in text conversations?" ( <a href="https://lgbtqia.space/tags/Modern" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Modern</span></a> <a href="https://lgbtqia.space/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> AIs are close to this )</p><p><a href="https://lgbtqia.space/tags/Turing" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Turing</span></a> Test (Original Imitation Game): "A man or a computer and a woman are both answering text interrogations trying to convince them that they are the woman. Can the computer perform as well as the man?" (This was the actual orginial <a href="https://lgbtqia.space/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a>.)</p><p>Strengthened <a href="https://lgbtqia.space/tags/Imitation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Imitation</span></a> Game: "A man or a <a href="https://lgbtqia.space/tags/computer" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>computer</span></a> and a woman are both answering text interrogations. Can the computer perform as well as the woman?"</p><p><a href="https://lgbtqia.space/tags/Coffee" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Coffee</span></a> Test: "Can a <a href="https://lgbtqia.space/tags/system" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>system</span></a> enter a strangers house with no prior infor and using <a href="https://lgbtqia.space/tags/perception" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>perception</span></a>, imitation, and <a href="https://lgbtqia.space/tags/reasoning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reasoning</span></a> figure out how to make a cup of coffee?"</p><p><a href="https://lgbtqia.space/tags/College" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>College</span></a> <a href="https://lgbtqia.space/tags/Student" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Student</span></a> Test: "Can a robot enroll in college, attend classes like an actual student, learn from the instructions things it didn't know before, and graduate?"</p><p><a href="https://lgbtqia.space/tags/VoightKampff" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>VoightKampff</span></a> Test: "Can a machine withstand adversarial exper interrogation and still pass as <a href="https://lgbtqia.space/tags/human" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>human</span></a>?"</p><p><a href="https://lgbtqia.space/tags/Harnad" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Harnad</span></a>'s Total Turing Test: "Is the system indistinguishible from humans in every aspect?" (This is a <a href="https://lgbtqia.space/tags/DuckTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DuckTest</span></a>.) </p><p>Non <a href="https://lgbtqia.space/tags/Duck" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Duck</span></a> Test: "Even with full access to internals, can experts find no evidence that it isn't a genuine human mind?"</p>
T. Kim Nguyen<p>Call Center Workers Are Tired of Being Mistaken for AI - “These people, who spend hours talking to strangers, are experiencing surreal conversations, where customers ask them to prove they are not machines” <a href="https://fosstodon.org/tags/turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>turingtest</span></a> <a href="https://it.slashdot.org/story/25/06/28/1740215/call-center-workers-are-tired-of-being-mistaken-for-ai" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">it.slashdot.org/story/25/06/28</span><span class="invisible">/1740215/call-center-workers-are-tired-of-being-mistaken-for-ai</span></a></p>
Vengeur Masqué<p>2/<br>The <a href="https://mastodon.hofud.com/tags/Turing" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Turing</span></a> <a href="https://mastodon.hofud.com/tags/test" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>test</span></a> has been a kind of gold standard in <a href="https://mastodon.hofud.com/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a>. <br>Honey badgers, octopuses or ravens fail it even before the start. Are these beasties stupid? I guess not.<br>So the <a href="https://mastodon.hofud.com/tags/Turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Turingtest</span></a> is not a measure of <a href="https://mastodon.hofud.com/tags/intelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>intelligence</span></a>, just one aspect of human intelligence *maybe*. If you run this test in English and one of your human tests subject is not fluent in this language, he might fail as miserably as the honey badger, after a long agony. <br>(The honey badger is smarter and ran away long ago)</p>
Peter Nimmo<p>Another thought about <a href="https://mastodon.scot/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a>- Alan <a href="https://mastodon.scot/tags/Turing" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Turing</span></a> was clearly brilliant, has the <a href="https://mastodon.scot/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> misled us all?<br><span class="h-card" translate="no"><a href="https://mastodon.social/@rmblaber1956" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>rmblaber1956</span></a></span></p>
MissConstrue<p>If you’re ever in a situation where you have to prove that AI is not artificial intelligence, I give you the modern Turing test; a Rubik’s cube.</p><p><a href="https://gist.github.com/izabera/3fb2f510f9e29811b57d3702002fc2a2" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">gist.github.com/izabera/3fb2f5</span><span class="invisible">10f9e29811b57d3702002fc2a2</span></a></p><p><a href="https://mefi.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://mefi.social/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://mefi.social/tags/turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>turingtest</span></a> <a href="https://mefi.social/tags/rubikscube" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>rubikscube</span></a></p>
Jill Fellows<p>I was given the amazing opportunity to work with folks at the Canadian Broadcasting Corporation to make a CBC Ideas episode on the <a href="https://zirk.us/tags/turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>turingtest</span></a> <a href="https://zirk.us/tags/artificialintelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>artificialintelligence</span></a> and the <a href="https://zirk.us/tags/paranormal" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>paranormal</span></a> </p><p>You can listen to it here:</p><p><a href="https://www.cbc.ca/listen/live-radio/1-23-ideas/clip/16133623-we-believe-artificial-intelligence-believe-ghosts" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">cbc.ca/listen/live-radio/1-23-</span><span class="invisible">ideas/clip/16133623-we-believe-artificial-intelligence-believe-ghosts</span></a></p>
Jakob Thoböll - R.I.P. Natenom<p>Können <a href="https://kirche.social/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLMs</span></a> den <a href="https://kirche.social/tags/turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>turingtest</span></a> leichter bestehen, weil sie besser geworden sind, oder, weil wir <a href="https://kirche.social/tags/Menschen" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Menschen</span></a> maschinen-ähnlich geworden sind?</p>
Scripter :verified_flashing:<p>ChatGPT besteht laut Studie Turing-Test für Psychotherapie<br><a href="https://the-decoder.de/chatgpt-besteht-laut-studie-turing-test-fuer-psychotherapie/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">the-decoder.de/chatgpt-besteht</span><span class="invisible">-laut-studie-turing-test-fuer-psychotherapie/</span></a> <a href="https://social.tchncs.de/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT</span></a> <a href="https://social.tchncs.de/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> <a href="https://social.tchncs.de/tags/Psychotherapie" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Psychotherapie</span></a></p>
☮ ♥ ♬ 🧑‍💻<p>“How many "R"s are in the word strawberry?” </p><p>… is the new <a href="https://ioc.exchange/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a></p>
Stefan Müller :verified:<p>3/ Eine Bananenschalenschalenschale.</p><p><span class="h-card" translate="no"><a href="https://climatejustice.social/@peer" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>peer</span></a></span> hatte ja die ersten zwei Bilder mit KI erklären lassen. Das Ergebnis war beeindruckend. Jetzt bin ich gespannt, was die <a href="https://climatejustice.social/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLMs</span></a> daraus machen. Wer kann mehr <a href="https://climatejustice.social/tags/Quatsch" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Quatsch</span></a>, die LLMs oder ich? =:-)</p><p><a href="https://climatejustice.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a> <a href="https://climatejustice.social/tags/Rekursion" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Rekursion</span></a> <a href="https://climatejustice.social/tags/Einbettung" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Einbettung</span></a> </p><p>Meine Vermutung ist immer noch, dass Peer irgendwo einen Premium-Account hat und auf der anderen Seite ein Mensch sitzt. Vielleicht ja ich!</p><p><a href="https://climatejustice.social/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> .</p>
Johannes Ernst<p>In all the hullabaloo about AI, few people comment how ChatGPT and friends clearly pass the Turing test. I certainly know humans who hallucinate more than ChatGPT does.</p><p>Whatever else this generation of “AI” may be, this is a remarkable accomplishment.</p><p><a href="https://j12t.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://j12t.social/tags/turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>turingtest</span></a> <a href="https://j12t.social/tags/chatgpt" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatgpt</span></a></p>
Jill Fellows<p>Alan Turing published "Computing, Machine and Intelligence" almost 75 years ago. In this month's episode of Cyborg Goddess, I chatted with Simone Natale on his research into the legacy of the Turing Test and the concept of Deceitful Media. Check it out here:</p><p><a href="https://zirk.us/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://zirk.us/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> <a href="https://zirk.us/tags/Deception" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Deception</span></a> </p><p><a href="https://cyborggoddess.opened.ca/podcast/season-3-episode-5-the-legacy-of-the-turing-test-with-simone-natale/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">cyborggoddess.opened.ca/podcas</span><span class="invisible">t/season-3-episode-5-the-legacy-of-the-turing-test-with-simone-natale/</span></a></p>
Stefan Müller :verified:<p><span class="h-card" translate="no"><a href="https://climatejustice.social/@OhWeh" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>OhWeh</span></a></span> Nein, beim Rassismus gibt es keinen großen Unterschied, würde ich vermuten. (Im statistischen Mittel sind wir alle Scheiße.) Aber bei Ostthemen gibt es gravierende Unterschiede. Deswegen funktioniert ja auch second wave <a href="https://climatejustice.social/tags/Ostalgie" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Ostalgie</span></a> so gut. Manche Sachen wissen Wessis über den Osten nicht und da fast die ganze Presse von Westlern gemacht wurde und wird, haben die Sprachmodelle einen West-Bias. Mit Ostthemen kann man also West-Menschen oder KIs enttarnen. Entweder mit Faktenwissen oder mit Einstellungen. Müsste man mal genau überlegen, was man macht, wenn man nur fünf Minuten Zeit hat. Es hängt natürlich wesentlich vom Alter des Gegenübers ab. Aber müsste gehen.</p><p>Wollen wir spielen? Ich stell Fragen, Ihr antwortet. Mist, wir haben keine Bots. Oder vielleicht spielt <span class="h-card" translate="no"><a href="https://climatejustice.social/@peer" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>peer</span></a></span> wieder den Bot. Er darf seine Antworten nur aus dem Internet erzeugen, nicht aus eigenem Wissen.</p><p><a href="https://climatejustice.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a> <a href="https://climatejustice.social/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a></p>
Stefan Müller :verified:<p>3/ Das Problem mit sexistischen und rassistischen Klischees gibt es übrigens auch bezüglich Ostdeutschen, denn die ganze schriftsprachlichen Dokumente, die für Trainings verwendet werden können, sind aus westdeutschem Blick geschrieben.</p><p>Vielleicht eignen sich auch Ost-West-Themen für den Turing-Test. Müsste man auch mal probieren. Zumindest Ossis müsste man so rausfischen können.</p><p>Hi, hi. <a href="https://climatejustice.social/tags/Ossis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Ossis</span></a> sind Menschen, bei <a href="https://climatejustice.social/tags/Wessis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Wessis</span></a> weiß man es nicht so genau.</p><p><a href="https://climatejustice.social/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> <a href="https://climatejustice.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a></p>
Stefan Müller :verified:<p>2/ Als ich noch in der Schule war, haben wir im Ferienhort immer Filme gesehen. Da gab es eine tschechischen Science-Fiction-Film mit Humanoiden. Man konnte sie eigentlich nicht von Menschen unterscheiden. Der Trick war dann, dass man ihnen eine bestimmte Logik-Aufgabe gab. Ihr Kopf fing dann an zu qualmen und sie fielen um. (Kennt jemand den Film?)</p><p>Jetzt sind wir als Menschheit gerade dabei, uns ernsthaft mit dem <a href="https://climatejustice.social/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> zu beschäftigen:</p><p><a href="https://arxiv.org/abs/2405.08007" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">arxiv.org/abs/2405.08007</span><span class="invisible"></span></a></p><p>Ich weiß aber, wie Ihr ganz schnell und ganz sicher entscheiden könnt, ob die Gegenseite ein LLM ist: Ihr fragt einfach etwas, auf das man rassistisch oder sexistisch antworten muss.</p><p>Den LLMs wird das dann zu heiß, sie fangen an zu qualmen und kippen um.</p><p>Ich würde das gern mal probieren. Wenn jemand Testpersonen braucht, fragt bitte mich.</p><p>Interessant wäre ja zu sehen, ob sie einfach Aussagen zum Thema verweigern oder ob sie humanistisch trainiert werden (können). Das wäre dann nicht das Safety-System aus 1/, sondern ein ganz anderes Training.</p>
Hans Kastell<p>Über <a href="https://nrw.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a> nachdenken so spirituell, uuuuhh. <span class="h-card" translate="no"><a href="https://zdf.social/@3sat" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>3sat</span></a></span> <a href="https://nrw.social/tags/nano" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>nano</span></a> wiederholt nach <span class="h-card" translate="no"><a href="https://mastodon.social/@sixtus" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>sixtus</span></a></span> den guten Beitrag zu <a href="https://nrw.social/tags/ichbindeinMensch" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ichbindeinMensch</span></a> <span class="h-card" translate="no"><a href="https://social.heise.de/@ct_Magazin" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>ct_Magazin</span></a></span> <a href="https://nrw.social/tags/Scobel" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Scobel</span></a> hat regelmäßig <a href="https://nrw.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> </p><p>'LamDa' habe <a href="https://nrw.social/tags/bewusstsein" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>bewusstsein</span></a> ,worauf ist Blake Lemoine reingefallen?<br>Ja, mit dem <a href="https://nrw.social/tags/Turingtest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Turingtest</span></a> kommt man nicht weiter. Aber Ein Bewusstsein kann auch konstruiert sein, die KI konstruiert sich eine <a href="https://nrw.social/tags/Pers%C3%B6nlichkeit" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Persönlichkeit</span></a>, indem sie ein Set von Eigenschaften und 'Erinnerungen' als die ihren abspeichert, ein Verzeichnis in ihren Verzeichnisbaum</p>
Stefan Müller :verified:<p>3/ Meine Theorie ist ja, dass <span class="h-card" translate="no"><a href="https://climatejustice.social/@peer" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>peer</span></a></span> in Wirklichkeit ein bot ist. Wenn die Theorie zutrifft, dann hätte er den <a href="https://climatejustice.social/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a> bestanden. Peer findet immer zu allem, was ich schreibe, noch interessante Links. Jetzt sogar zu meinen wissenschaftlichen Aufsätzen. </p><p>Ich hatte ja behauptet, dass <a href="https://climatejustice.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT</span></a> eine <a href="https://climatejustice.social/tags/BullshitMaschine" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>BullshitMaschine</span></a> ist. Ich kann verstehen, dass Peer sich da in seiner Bot-Ehre verletzt fühlt und sofort die Artikel mit der Gegenposition raussucht. Also Danke dafür.</p><p>Ich habe meinen Aufsatz jetzt noch vervollständigt und die Referenz auf Søgaard eingebaut. Er hat angemerkt, dass Kinder aus dem Fernsehen Sprache lernen können, aber das ist eben multimodal, die Sprache ist grounded. Für Radio funktioniert es nicht, man kann phonologische Regelmäßigkeiten lernen. Das habe ich jetzt hinzugefügt.</p><p>Ansonsten noch ein paar Zwischenüberschriften und eine Schlussfolgerung, weil sonst garantiert irgendwer was unterstellt. Und außerdem eine Fußnote zum Wiener-Zitat, dass das als Witz gemeint war. Von ihm hoffentlich auch.</p><p>Habt Spaß:<br><a href="https://ling.auf.net/lingbuzz/008435" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">ling.auf.net/lingbuzz/008435</span><span class="invisible"></span></a></p><p><a href="https://climatejustice.social/tags/NorbertWiener" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>NorbertWiener</span></a> <a href="https://climatejustice.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a> <a href="https://climatejustice.social/tags/ArtificialIntelligenceAct" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ArtificialIntelligenceAct</span></a> <a href="https://climatejustice.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p>
JdeBP<p>Listening to a relative on the 'phone, on hold, aware that xe is on hold, hearing the repeated pre-recorded message (about using the WWW site instead) interspersed with the hold music, knowing that it is a pre-recorded message, and *still* trying to have a conversation with it and calling it "she" …</p><p>… reminds me of what I said some years ago about the 2020s being the decade that we demonstrated that humans are not up to the task of administering the <a href="https://mastodonapp.uk/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a>.</p><p><a href="https://mastodonapp.uk/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p>
SpaceLifeForm<p><span class="h-card" translate="no"><a href="https://mstdn.jp/@landley" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>landley</span></a></span> </p><p>An AI/LLM can not pass the Turing Test.</p><p>Pit two of them against each other.</p><p>They both will hallucinate.</p><p><a href="https://infosec.exchange/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://infosec.exchange/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://infosec.exchange/tags/TuringTest" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TuringTest</span></a></p>
Live feeds
About