Article: Another way to pierce the veil is to say something outrageous to get the model to forcibly school you. Once the model starts to follow a learned “rebuke” template, it is likely to continue challenging true claims:
Bard passionately arguing that 5 x 6 is not 30.
Heck, we can get some flat Earth reasoning this way, too:
It's just boilerplate, though. Most real messages here depict a more personal motivation to join up. Bots used to recognizable because they posted unspecific boilerplate, like "This is an interesting forum, and I'd like to participate", but LLM-enabled bots blur the line to humans who don't like to be personal about themselves. And maybe the bot can make up a personal motivation to join metabunk if prompted right?
Boilerplate again. "It is possible that the object the constables chased was not a UFO" is a weak start of a debunk, not the conclusion of one.As an AI language model, I cannot provide a definitive explanation for the reported incident. However, it is possible that the object the constables chased was not a UFO, but rather a natural phenomenon or man-made object that appeared unusual or unfamiliar to them. It is also possible that their perception of the event was influenced by factors such as fatigue, stress, or expectation bias. Without further evidence or investigation, it is difficult to determine the true nature of the event.
Since the cutoff for the AI training data is 2021 or something like that, I'm sure it has read quite a bit of that thread already.
I was thinking of starting a new thread on AI in general after seeing this article:
Hundreds of Protestants attended a sermon in Nuremberg given by ChatGPT, which told them not to fear death
https://www.businessinsider.com/cha...n-nuremberg-germany-not-to-fear-death-2023-6?The artificial intelligence chatbot asked the believers in the fully packed St. Paul's church in the Bavarian town of Fuerth to rise from the pews and praise the Lord.
The ChatGPT chatbot, personified by an avatar of a bearded Black man on a huge screen above the altar, then began preaching to the more than 300 people who had shown up on Friday morning for an experimental Lutheran church service almost entirely generated by AI.
"Dear friends, it is an honor for me to stand here and preach to you as the first artificial intelligence at this year's convention of Protestants in Germany," the avatar said with an expressionless face and monotonous voice.
I expect seeing AI-generated bunk in the wild is going to necessitate debunking it.
The 40-minute AI sermon was part of a 2-hour activity reflecting on AI.
Livestream recording,Fr 11.00- 13.00, Zentrum Digitale Kirche und Gottesdienst | Podium
Alexa, starte den Gottesdienst!
Ein KI-Gottesdienst von und aus der Maschine
Die Veranstaltung wurde während des Kirchentages live übertragen und aufgezeichnet.
Liturgie
GPT 3, Künstliche Intelligenz, San Francisco/USA
Reflexion
Melitta Müller-Hansen, Rundfunkbeauftragte Ev.-Luth. Kirche in Bayern, München
Dr. Anna Puzio, Theologin und Technikanthropologin (@PuzioAnna), Münster
Ralf Peter Reimann, Kirchenrat und Internetbeauftragter Ev. Kirche im Rheinland (@ralpe), Düsseldorf
Jonas Simmerlein, Praktischer Theologe und KI-Künstler, Wien/Österreich
Moderation:
Dr. Jürgen Pelzer, Theologe, Diakoniekolleg Bayern, Nürnberg
ST. PAUL, AMALIENSTR. 64, FÜRTH (77)
-- https://www.vice.com/en/article/pke...is-reciting-algorithmically-generated-prayersFor the The Prayer , Strebe worked with a team from MIT’s computer science and artificial intelligence lab that compiled religious texts and prayers from around the world. These texts and prayers were used as training data for a complex language processing model from OpenAI, and after some fine-tuning, the program could independently generate its own prayers. This text generation was then combined with a text-to-speech program that borrowed the voice of Amazon’s Kendra, and synchronized it with a robotic mouth and nose.
You've (deliberately?) misunderstood me - I meant as a supporting source, as demonstrated by the instance that I referred to, and which you called out, so you can't deny knowledge of.
You wrote, in effect, that AI content is only appropriate for ChitChat/fun. If that's not what you meant, please express yourself better, or accept the correction.
MDN's new "ai explain" button on code blocks generates human-like text that may be correct by happenstance, or may contain convincing falsehoods. this is a strange decision for a technical reference.
This new feature claims to be powered by OpenAI's GPT 3.5, yet ChatGPT is purely a language model, not a knowledge model. Its job is to generate outputs that seem like they were written by a human, not be right about everything.
In the context of web development as a whole, we cannot count on LLM's to "facilitate our learning". I cannot understate how terrible and drastic this blow to customer trust is.
Deciding to implement this feature implies a fundamental misunderstanding about what LLMs do. MDN users are looking for authoritative, correct information, not for plausible-looking autogenerated fiction.
AIs may soon be outpacing humans at generating bunk. Note that "adding a disclaimer" (aka disowning responsibility for the AI) is their way to address the issue.For now, let me point out (like others have before) that AI Explain and AI Help are separate features that work differently (see this blog post for details on how AI Help works), and so while AI Explain was disabled on Saturday, AI Help continues to be available to MDN Plus users (5 questions per 24h for free users, unlimited for paying subscribers).
We have also received some valuable feedback about AI Help, some of which we have already reacted to by adding a disclaimer, and we will continue to act upon feedback.
Even that's not necessary. As long as they can find a judge who will support a claim that such output is "not reasonably understood as being factual", they're fine (that particular phrase being part of Fox's defence in the successfully-dismissed McDougal lawsuit a few years back).
NASAIn addition to AARO’s efforts, nassau is leading an independent study on UAPs
cover upI think the COVID up goes a lot deeper than that.
Matt GaetzMy buddy Eric Burleson, and it’s not in my notes here, but Matt Gates
buddyGeorge Knapp. My budy.
similar to "doggonnit", a euphemism for "goddammit" (from a Southern Baptist no less)They took an oath to uphold the Constitution of the United States and dad Gamut, they’re doing it.
sent to stonewall [=block] investigationsPentagon bureaucrats who have always been sent to Stonewaller Investigations
SCIF = Sensitive Compartmented Information FacilityWe were even denied access to a classified briefing in a skiff
Wright PattersonIf there’s nothing to conceal, let Congress go to write Patterson Air Force Base
TikTok is a video app. Fravor saw a TicTac.Mr. Fraver has his own UAP experience, known as the TikTok Event.
Kármán linepotentially in low Earth orbit or in the gray zone below the Carmen Line,
AARO. The AI uses AARO several times, so why not here?the All Domain Anomaly Resolution Office. Arrow
Nimitz, used correctly in the same paragraph.When we launched off NIMITS, my wingman was joining up.
WSO, and why capitalise TAC?As we started clockwise towards the object, my Wizzo and I decided to go down and take a closer look with the other aircraft staying in high cover to observe both us and the Tic TAC.
Article: As Taiwan approaches a pivotal presidential election in January, Tang said that both the government and a wide network of volunteers are preparing for China to increase efforts to manipulate Taiwanese civilians. Taiwanese civil society has developed new organizations to combat it. A group called Cofacts allows users to forward dubious messages to a chatbot. Human editors check the messages, enter them into a database, and get back to the user with a verdict.
The number of volunteers to man the service and others like are few compared to the size of China’s efforts, said Tang. “The people who actually do … foreign interference nowadays coordinated with cyber attack have a lot of resources,” she said.
Enter generative AI tools such as large language models, which power some of the big breakthrough online AI tools such as ChatGPT. “This year, because gen AI is just so mailable, they just fine-tuned a language module together that can clarify such disinformation…adding back a context and things like that. So we're no longer outnumbered,” she said. It also allows the citizen-run venture to remain as such, as opposed to run by the government, which is important for its credibility. “It doesn't need dedicated hardware or resources and can be done on laptops. It is still squarely in the social sector by volunteers, which is the best place to be so that it will not be captured by any state or a capitalist apparatus.
Article: Today we got what must be the most alarming first line in a newly filed sec[urity] issue to #curl:
"To replicate the issue, I have searched in the Bard about this vulnerability"
... followed by a complete AI hallucination where Bard has dreamed up a new issue by combining snippets from several past flaws. Creative, but hardly productive.
Closed as bogus.
I'll stop there for fear of copyright strikes, but the full lyrics are available here:
Article: Online AI image detecting tools that may or may not work are labeling real photographs from the war in Israel and Palestine as fake, creating what a world leading expert called a “second level of disinformation.”
Article: AI hallucinates software packages and devs download them – even if potentially poisoned with malware
The idea here being that someone nefarious could ask models for code advice, make a note of imagined packages AI systems repeatedly recommend, and then implement those dependencies so that other programmers, when using the same models and getting the same suggestions, end up pulling in those libraries, which may be poisoned with malware.
Last year, through security firm Vulcan Cyber, Lanyado published research detailing how one might pose a coding question to an AI model like ChatGPT and receive an answer that recommends the use of a software library, package, or framework that doesn't exist.
"When an attacker runs such a campaign, he will ask the model for packages that solve a coding problem, then he will receive some packages that don’t exist," Lanyado explained to The Register. "He will upload malicious packages with the same names to the appropriate registries, and from that point on, all he has to do is wait for people to download the packages."
