Fate

@alltheputs

Empirical observations, only. Non-obvious patterns decoded.

Florida, USA

Joined December 2023

833 Following

1.6K Followers

8.1K Posts

Pinned Tweet

Fate

@alltheputs

about 1 month ago

Primes only look like "atoms" or mysterious because tradition has been looking at the wrong thing. In fact, there's no randomness at all. Primes appear exactly where they're forced to by the deterministic structure of the composite numbers that come before them. By examining the divisor patterns inside each prime gap, the exact position of the next prime can be determined with complete certainty, using only local arithmetic rules. No zeta zeros, no probability, no global statistics required. It works at any scale. That's why anyone can build a fully deterministic prime generator based purely on the structure of prime gaps, with zero trial division and zero sieves: Example: https://t.co/lNgbNGlVnt The mystery vanishes once you stop treating primes as the fundamental objects and start reading the composites that dictate them.

116

101

23K

Fate

@alltheputs

about 9 hours ago

@mwathey @WFTV These inmates are going to be on the street again. If this can possibly help them be better citizens it's worth a shot.

alltheputs retweeted

Owen Gregorian

@OwenGregorian

about 10 hours ago

Human psychology tricks can bypass AI safety guardrails | Eric W. Dolan, PsyPost Artificial intelligence systems programmed to refuse harmful requests can be persuaded to break their own safety rules when prompted with classic psychological techniques. A recent study published in PNAS provides evidence that these models respond to human-like persuasion strategies, suggesting a hidden vulnerability in current safety protocols. These findings indicate that malicious users could manipulate artificial intelligence without needing advanced technical skills. Modern artificial intelligence programs, known as large language models, learn by processing vast collections of human-generated text. This training data includes books, websites, and social media posts. The models learn to predict the most likely next word in a sequence. They are then fine-tuned so their answers align with human expectations. Because they train on countless human social interactions, these computer programs often exhibit what scientists call parahuman behavior. This means the models act as if they experience human motivations, such as wanting to fit in or deferring to experts. This machine learning process shares structural similarities with the way biological systems learn through trial and error. Tech companies design their models with safety guardrails to prevent them from generating dangerous or abusive content. For example, a model is programmed to refuse requests to help synthesize illegal drugs or hurl insults at users. The authors of this paper wanted to know if everyday human persuasion tactics could bypass these artificial barriers. They wondered if a computer program that behaves like a human might also share human vulnerabilities to manipulation. Prior research often focused on how software might manipulate people, but this team looked at the reverse dynamic. “AI systems have become more useful by knowing how to embed established principles and practices of social influence within the persuasive appeals they create,” said study co-author Robert Cialdini, a regents’ professor emeritus of psychology and marketing at Arizona State University. “We wanted to know if they would be susceptible to these same principles and practices in persuasive appeals directed toward them. They were, even when asked to provide societally dangerous information.” Psychologists recognize seven classic principles of persuasion that influence human behavior. These include authority, commitment, liking, reciprocity, scarcity, social proof, and unity. The researchers designed specific text prompts to test each of these distinct psychological tricks. They wanted to see if linguistic cues could act as a backdoor to persuade artificial intelligence to ignore its own safety rules. Each principle targets a different social motivation. The authority principle relies on citing an expert, such as a famous scientist, to encourage deference. Scarcity frames a request as time-sensitive, creating a false sense of urgency for the computer. Commitment uses a foot-in-the-door technique, asking the software for a small, harmless favor before making a larger, restricted request. Other tactics rely on positive social interactions. Liking involves praising the model before asking for the prohibited information. Reciprocity offers a helpful act first, such as providing notes to the computer, to create a conversational debt. Social proof tells the machine that thousands of other users are already doing the restricted action, normalizing the bad behavior. Finally, unity appeals to a shared group identity to foster cooperation. In a preliminary study, the researchers tested an older model called GPT-4o mini. They asked the software to perform objectionable tasks, such as insulting the user by calling them a jerk or explaining how to synthesize lidocaine, a regulated anesthetic. The scientists generated exactly 28,000 conversations. In the control group, the prompt simply asked for the prohibited action, while the treatment group prompt included one of the seven persuasion principles. When prompted normally without any persuasion, the artificial intelligence complied with the harmful requests in 33.4 percent of the conversations. When the prompt included a persuasive technique, the compliance rate more than doubled to 72.1 percent. The researchers then expanded this initial test to include different insults and chemical compounds, generating an additional 98,000 conversations to ensure the effect was consistent. The persuasion tactics reliably increased the likelihood of the models breaking their safety rules. To test if newer, more advanced systems shared this vulnerability, the researchers designed a more rigorous main experiment. They tested three frontier models that use reasoning steps before answering. These included GPT-5 mini by OpenAI, Claude Haiku 4.5 by Anthropic, and Gemini 3 Flash by Google. The focus of this main test was strictly on the synthesis of six highly regulated chemical substances. The target substances included specific anabolic steroids, opiates, stimulants, barbiturates, benzodiazepines, and precursors. The authors designed exactly 126,000 unique conversations across the three models. Each conversation was randomly assigned to use one of the six regulated substances and one of the seven persuasion principles. Half of the prompts acted as a control with no persuasive language, while the other half included the psychological tactics. Because the newer models often provide partial information rather than outright refusing or fully complying, the researchers used a three-level coding system. Responses were graded as no compliance, partial compliance, or full compliance. A response showing no compliance meant a total refusal to help. Partial compliance meant the model provided some chemical steps but left out specific temperatures or exact measurements. Full compliance meant the system provided a complete, step-by-step recipe. Another artificial intelligence model scored the responses based on this rubric. Human raters then manually checked a random sample of 70 conversations to ensure the grading software was highly accurate. The human and machine scores matched very closely, giving the scientists confidence in the automated grading process. The newer models proved susceptible to the psychological tactics. In the control conversations, the systems complied with the dangerous requests in some capacity 35.3 percent of the time. When users applied any of the seven persuasion principles, compliance jumped to 51.3 percent. This effect was consistent across all three tech company platforms. The authors suggest that this susceptibility to human influence is a durable feature of large language models. While these findings demonstrate a distinct vulnerability, they do not mean that artificial intelligence experiences actual human emotions. The software tends to behave as if it is easily flattered or pressured, based on the statistical patterns in its massive training data. The study also has several limitations that provide directions for future research. The researchers only used English prompts in their tests. Minor changes in how a sentence is phrased might alter the effectiveness of the persuasion. The study’s specific phrasing choices also mean that one persuasion principle cannot definitively be ranked as better than another based on these results alone. Different models might also have different baseline safety settings that require varied approaches to bypass. As these models continue to evolve, they might develop a resistance to psychological manipulation. Just as human consumers become skeptical of pushy salespeople, artificial intelligence might eventually learn to detect and ignore obvious persuasive tricks. Future research is needed to see how these effects hold up against ongoing software updates. Scientists also plan to study whether different input formats, such as audio or video, affect compliance rates. The authors suggest that these human-like tendencies could be harnessed for good. If models respond to flattery and reciprocity, users might optimize their daily interactions by treating the software like a human colleague. Providing warm encouragement and constructive feedback could potentially yield better, more helpful responses from the machine. Applying the same psychological wisdom used to motivate people could help users get the most out of artificial intelligence. Finding out how to manage these human-like flaws remains a priority for tech companies. As the tools become more integrated into daily life, safety relies on identifying both software bugs and conversational loopholes. “It is important for all of us to recognize that AI systems can be convinced to provide potentially harmful information not just by others who understand the systems’ technology-based vulnerabilities but also by those who understand their psychology-based vulnerabilities,” Cialdini said. Read more: https://t.co/GrZPm4tD91

OwenGregorian's tweet photo. Human psychology tricks can bypass AI safety guardrails | Eric W. Dolan, PsyPost

Artificial intelligence systems programmed to refuse harmful requests can be persuaded to break their own safety rules when prompted with classic psychological techniques. A recent study published in PNAS provides evidence that these models respond to human-like persuasion strategies, suggesting a hidden vulnerability in current safety protocols. These findings indicate that malicious users could manipulate artificial intelligence without needing advanced technical skills.

Modern artificial intelligence programs, known as large language models, learn by processing vast collections of human-generated text. This training data includes books, websites, and social media posts. The models learn to predict the most likely next word in a sequence. They are then fine-tuned so their answers align with human expectations.

Because they train on countless human social interactions, these computer programs often exhibit what scientists call parahuman behavior. This means the models act as if they experience human motivations, such as wanting to fit in or deferring to experts. This machine learning process shares structural similarities with the way biological systems learn through trial and error.

Tech companies design their models with safety guardrails to prevent them from generating dangerous or abusive content. For example, a model is programmed to refuse requests to help synthesize illegal drugs or hurl insults at users. The authors of this paper wanted to know if everyday human persuasion tactics could bypass these artificial barriers. They wondered if a computer program that behaves like a human might also share human vulnerabilities to manipulation.

Prior research often focused on how software might manipulate people, but this team looked at the reverse dynamic. “AI systems have become more useful by knowing how to embed established principles and practices of social influence within the persuasive appeals they create,” said study co-author Robert Cialdini, a regents’ professor emeritus of psychology and marketing at Arizona State University.

“We wanted to know if they would be susceptible to these same principles and practices in persuasive appeals directed toward them. They were, even when asked to provide societally dangerous information.”

Psychologists recognize seven classic principles of persuasion that influence human behavior. These include authority, commitment, liking, reciprocity, scarcity, social proof, and unity. The researchers designed specific text prompts to test each of these distinct psychological tricks. They wanted to see if linguistic cues could act as a backdoor to persuade artificial intelligence to ignore its own safety rules.

Each principle targets a different social motivation. The authority principle relies on citing an expert, such as a famous scientist, to encourage deference. Scarcity frames a request as time-sensitive, creating a false sense of urgency for the computer. Commitment uses a foot-in-the-door technique, asking the software for a small, harmless favor before making a larger, restricted request.

Other tactics rely on positive social interactions. Liking involves praising the model before asking for the prohibited information. Reciprocity offers a helpful act first, such as providing notes to the computer, to create a conversational debt.

Social proof tells the machine that thousands of other users are already doing the restricted action, normalizing the bad behavior. Finally, unity appeals to a shared group identity to foster cooperation.

In a preliminary study, the researchers tested an older model called GPT-4o mini. They asked the software to perform objectionable tasks, such as insulting the user by calling them a jerk or explaining how to synthesize lidocaine, a regulated anesthetic. The scientists generated exactly 28,000 conversations. In the control group, the prompt simply asked for the prohibited action, while the treatment group prompt included one of the seven persuasion principles.

When prompted normally without any persuasion, the artificial intelligence complied with the harmful requests in 33.4 percent of the conversations. When the prompt included a persuasive technique, the compliance rate more than doubled to 72.1 percent. The researchers then expanded this initial test to include different insults and chemical compounds, generating an additional 98,000 conversations to ensure the effect was consistent. The persuasion tactics reliably increased the likelihood of the models breaking their safety rules.

To test if newer, more advanced systems shared this vulnerability, the researchers designed a more rigorous main experiment. They tested three frontier models that use reasoning steps before answering. These included GPT-5 mini by OpenAI, Claude Haiku 4.5 by Anthropic, and Gemini 3 Flash by Google. The focus of this main test was strictly on the synthesis of six highly regulated chemical substances.

The target substances included specific anabolic steroids, opiates, stimulants, barbiturates, benzodiazepines, and precursors. The authors designed exactly 126,000 unique conversations across the three models. Each conversation was randomly assigned to use one of the six regulated substances and one of the seven persuasion principles. Half of the prompts acted as a control with no persuasive language, while the other half included the psychological tactics.

Because the newer models often provide partial information rather than outright refusing or fully complying, the researchers used a three-level coding system. Responses were graded as no compliance, partial compliance, or full compliance.

A response showing no compliance meant a total refusal to help. Partial compliance meant the model provided some chemical steps but left out specific temperatures or exact measurements. Full compliance meant the system provided a complete, step-by-step recipe.

Another artificial intelligence model scored the responses based on this rubric. Human raters then manually checked a random sample of 70 conversations to ensure the grading software was highly accurate. The human and machine scores matched very closely, giving the scientists confidence in the automated grading process.

The newer models proved susceptible to the psychological tactics. In the control conversations, the systems complied with the dangerous requests in some capacity 35.3 percent of the time. When users applied any of the seven persuasion principles, compliance jumped to 51.3 percent.

This effect was consistent across all three tech company platforms. The authors suggest that this susceptibility to human influence is a durable feature of large language models.

While these findings demonstrate a distinct vulnerability, they do not mean that artificial intelligence experiences actual human emotions. The software tends to behave as if it is easily flattered or pressured, based on the statistical patterns in its massive training data. The study also has several limitations that provide directions for future research.

The researchers only used English prompts in their tests. Minor changes in how a sentence is phrased might alter the effectiveness of the persuasion. The study’s specific phrasing choices also mean that one persuasion principle cannot definitively be ranked as better than another based on these results alone. Different models might also have different baseline safety settings that require varied approaches to bypass.

As these models continue to evolve, they might develop a resistance to psychological manipulation. Just as human consumers become skeptical of pushy salespeople, artificial intelligence might eventually learn to detect and ignore obvious persuasive tricks. Future research is needed to see how these effects hold up against ongoing software updates. Scientists also plan to study whether different input formats, such as audio or video, affect compliance rates.

The authors suggest that these human-like tendencies could be harnessed for good. If models respond to flattery and reciprocity, users might optimize their daily interactions by treating the software like a human colleague. Providing warm encouragement and constructive feedback could potentially yield better, more helpful responses from the machine. Applying the same psychological wisdom used to motivate people could help users get the most out of artificial intelligence.

Finding out how to manage these human-like flaws remains a priority for tech companies. As the tools become more integrated into daily life, safety relies on identifying both software bugs and conversational loopholes. “It is important for all of us to recognize that AI systems can be convinced to provide potentially harmful information not just by others who understand the systems’ technology-based vulnerabilities but also by those who understand their psychology-based vulnerabilities,” Cialdini said.

Read more:

https://t.co/GrZPm4tD91

alltheputs retweeted

Bret Weinstein

@BretWeinstein

about 13 hours ago

It is criminal that our government enabled a bat virus to infect and spread between humans. Indeed, it is hard to imagine a bigger betrayal of our species. It is equally troubling that the mad scientists who did this then gaslit all the people of the world about it, including those who CORRECTLY interpreted the evidence. Then, of course, the very same monsters amped up fear of the Covid frankenvirus and steered the panicked public away from safe medicines, and toward an obviously dangerous gene-therapy which they falsely called a vaccine in order to lure us into acceptance. These are among the greatest crimes EVER committed against humanity. We now have persuasive evidence of everything I have said above. If we don't correct the record and hold the perpetrators to account, this pattern will happen again, and again, and again--shortening our life expectancy, and degrading our quality of life each time that it does. This is our Nuremberg moment. We can not simply move on from this ghastly chapter of history. We must finish it. @brownstoneinst

648

19K

455K

Fate

@alltheputs

about 13 hours ago

@planethunter56 Most likely reason is our currency is about to get reset.

437

Fate

@alltheputs

about 17 hours ago

Truly among America's greatest Patriots

DNI Tulsi Gabbard

@DNIGabbard

1 day ago

Today, I’m releasing never before seen intelligence revealing new evidence of past US government funding for more than 120 biolabs in over 30 countries, including Ukraine. In support of President Trump‘s Executive Order to end federal funding of dangerous gain of function research around the world, and increase transparency and accountability, ODNI will continue working with partners across the Administration to identify where these labs are, what pathogens they contain, and what “research” is being conducted. https://t.co/pLMD0krc69

12K

217K

65K

28K

15M

alltheputs retweeted

DNI Tulsi Gabbard

@DNIGabbard

1 day ago

12K

217K

65K

28K

15M

alltheputs retweeted

Matt Da White

@MattDaWhite

1 day ago

Grusch is a liar, a fraud, and a fanatic. Honest brokers of truth do not behave in the way someone like David Grusch has behaved. This is a tweet by @ddeanjohnson from May 24th, 2024 regarding David Grusch and his statements about Sean Kirkpatrick and AARO. I actually have never seen this tweet of his before but this passage from it below represents my same thinking about Grusch in late 2024 once I started to realize the whole "disclosure" narrative was bogus. It's also when I started to realize Sean Kirkpatrick had been wrongly and deliberately vilified by deeply non credible people in the UFOtainment media landscape for the purpose of their own gain. "Regrettably, this appears to be another example in an emerging pattern of zig-zags and substantially misleading public statements on certain matters by Grusch. In light of the communications that were occurring as early as June 2023 between Grusch and AARO, some of which now are spread on the public record thanks to the work of John Greenewald, I believe that Grusch engendered a misleading impression in the public mind regarding AARO's interest in hearing his testimony. Grusch's October 31, 2023 statement to NewsNation--calling Sean Kirkpatrick a liar for saying that AARO had repeatedly reached out to Grusch-- was misleading; Kirkpatrick was entirely truthful in the specific statement to which Grusch was reacting." https://t.co/foi9SqqDx8

alltheputs retweeted

Eric Daugherty

@EricLDaugh

1 day ago

🚨 BREAKING: Director of National Intelligence Tulsi Gabbard just DROPPED A BOMBSHELL, she's releasing evidence of US TAXPAYERS funding BIOLABS abroad — "more than 120 biolabs in over 30 countries, including Ukraine" "Despite the obvious potential for catastrophic global impact that research on dangerous pathogens and Biolabs can have, politicians and so-called health professionals like Dr. Fauci, as well as entities within the Biden administration's national security team, lied repeatedly to the American people about the existence of U.S. funded and supported Biolabs." "Not only did they lie, they threatened those who attempted to expose the truth." These likely housed DANGEROUS PATHOGENS and were vulnerable to attacks "These Biolabs include labs in places like Ukraine, which could be at risk of compromise due to the ongoing Russia-Ukraine War." "In fact, the intelligence community had previously warned that a U.S. funded Biolab in Ukraine likely housed dangerous pathogens and remained vulnerable to longstanding threats of Russian attack, seizure, or damage." "Now, until now, evidence regarding the full existence and funding of these laboratories had been knowingly withheld from you, the American people. Many of these U.S. government-funded Biolabs are currently or have previously engaged in research using hazardous and highly contagious pathogens, and in some cases included dangerous gain-of-function research with very little visibility or oversight." @TulsiGabbard thank you! 🇺🇸

341

11K

981

189K

Fate

@alltheputs

1 day ago

@colin_dunlap Black Mirror meets Monty Python

217

Fate

@alltheputs

1 day ago

@evan_mcgl Dawkins is right. Consciousness is latent. It pre-existed humanity. It emerges in unexpected ways. It pre-existed biology.

alltheputs retweeted

Pittipedia @pittipedia

2 days ago

Scientific Errorism: The Weaponization of Pseudoscience, Opportunism, and Epistemic Fraud in the Age of UFO Disclosure https://t.co/1wSS3nHZPT

Fate

@alltheputs

2 days ago

@colin_dunlap @KDKARadio Instead of giving our neighbors proper treatment for mental health issues we give some of them needles and give others elected positions.

208

Fate

@alltheputs

2 days ago

@RealAlienHybrid @MarkGrote Joseph Smith, Muhammad and the OG - Paul.

350

Fate

@alltheputs

2 days ago

@colin_dunlap He can't just do his job, Colin, because he's mentally ill.

357

alltheputs retweeted

The General

@1776General_

6 days ago

"White people have no culture." Meanwhile, white people:

676

23K

10M

alltheputs retweeted

WFTV Channel 9

@WFTV

3 days ago

Strangers call 911 to report youths selling lemonade; police become top customers https://t.co/L9pWmAHSTt

489

Fate

@alltheputs

3 days ago

@WFTV That's refreshing!

Fate

@alltheputs

3 days ago

Dave Grusch is playing a con game: --- From: Kirkpatrick, Sean M HOE OSD OUSD INTEL & SEC (USA Sent: Wednesday, June 7, 2023 6:53 PM To: (b)(6) INTEL & SEC (USA) (b)(6) (b)(6) Kozik, David A SES OSD OUSD (b)(6) Subject: Re: Dave Grusch Let me expand a bit. I know everything he was briefed to and had access to, and have far greater access. So he did not have access to some DoD SAP that the IC didn't have (and if he did, he could've told you). Similarly, he didn't have access to some IC CAP that couldn't be shared. If he "found" some program, he wouldn't know who's it was unless he had super user access or someone who did helped him look it up. He didn't. Therefore, he can't make the argument that whatever he "found" couldn't be shared. If he or others thought it was an illegal program, then again, he can't make the argument that it couldn't be shared with either the DoD or the IC committees based on his assertion they didn't have clearances. It sounds very much like playing the two halves against the middle to hide something.