Brane Space: "DebunkBot" Chat GPT-4 Debunked Trying To Knock Down JFK Assassination Conspiracy

Also missed by Chatbot: how angle of shot in rear had to be manipulated

And Oswald being killed to negate his testimony

According to latest news out of academia, a 'conversation' with a chatbot nicknamed 'Debunkbot' "lasting under 10 minutes mellowed down" conspiracy theorists. Including those believing some of the most deeply entrenched conspiracies, such as those involving the COVID-19 pandemic, a new study has found. A team, including researchers from Cornell, American University and the Massachusetts Institute of Technology, designed the AI-model ChatGPT to be highly persuasive and engage participants in tailored dialogues, such that participants' conspiratorial claims were factually rebutted by the chatbot.

For this, the researchers leveraged advancements in GPT-4, the large language model powering ChatGPT, that "has access to vast amounts of information and the ability to generate bespoke arguments." A form of artificial intelligence, a large language model is trained on massive amounts of textual data and can, therefore, respond to users' requests in the natural language.

The team found that among over 2,000 self-identified conspiracy believers, conversations with the AI-model ChatGPT lowered the average participant's beliefs by about 20 per cent. Further, about a fourth of the participants, all of whom believed a certain conspiracy previously, "disavowed" it after chatting with the AI-powered bot, the researchers said. The study authors wrote in the study published in the journal Science.

"Large language models can thereby directly refute particular evidence each individual cites as supporting their conspiratorial beliefs,"

In two separate experiments, the participants were asked to describe a conspiracy theory they believed in and provide supporting evidence, following which they engaged in a conversation with the chatbot. The conversation lasted 8.4 min on average and comprised three rounds of back-and-forth interaction (not counting the initial elicitation of reasons for belief from the participant), ..." the authors wrote. A control group of participants discussed an unrelated topic with the AI.

The findings suggested that ''any type of belief that people hold that is not based in good evidence could be shifted,'' said study co-author Gordon Pennycook, a cognitive psychologist at Cornell University. Adding:

''It's really validating to know that evidence does matter,''

The researchers began by asking Americans to rate the degree to which they subscribed to 15 common conspiracy theories, including that the virus responsible for COVID-19 was created by the Chinese government. Three examples given in the study and published in a mainstream media site Sunday were as follows:

The first and last passed the AI-chatbot competency test, but the middle effort - dealing with Lee Oswald- failed. The reasons are: a) The rifle allegedly used was a stage prop, a fake, confirming Oswald never made the shots, period, and (b) The Warren Commision's own team of crack shooters was unable to make the shots even after rebuilding the prop rifle.

In respect of (b), the rifle sight itself was rebuilt and “metal shims were fitted to provide a degree of accuracy previously absent’. When Ronald Simmons, the Chief of the Infantry Weapons Evaluation Branch of the Army’s Ballistics Research Division was asked about this he replied: “Well, they could not sight the weapon in using the telescope(#). Adding that the aiming apparatus had to be rebuilt by a machinist with two shims added, one to adjust for the elevation, the other for the azimuth.

Even with these changes, just one of the three expert riflemen was able to get off three shots in under 5.6 seconds – the designated time interval for total shots declared by the Warren Commission. And most to the point: none of the total 18 shots fired struck the targets (ibid.). in the head or the neck. In other words, from a technical standpoint of duplicating the alleged Oswald shots- this trio of experts failed.

In respect of point (a): When researcher Patricia Dumais - who suspected the rifle was a stage prop and wanted to see the internal control number - was informed by the National Archives (H.E. Livingstone, 1993, 'Killing the Truth', p. 204.):

'We cannot disassemble Lee Harvey Oswald's rifle because this action might be destructive to the object.’'

Beyond that, the 'genius' bot never processed that the angle of shots had been manipulated (e.g. by the WC Commissioner Gerald Ford) to reduce the total number to remove conspiracy from consideration, e.g.

Brane Space: Frequently Asked Questions on the JFK Assassination (Pt. 5):The Bullets, Wounds (brane-space.blogspot.com)

Oh yeah, and Oswald himself was snuffed by gangster affiliated Jack Ruby before he could testify before the Warren Commission, e.g

The Southern Poverty Law Center: Still In Over Its Head On the JFK Assassination

So all the Warren Commission's devout followers have been snookered for nearly 60 years by a colossal hoax. And we can now add Chat GPT-4 to that list. But the failure wasn't a one off. Not by any means.

The bot also tackled the “theory” that the U.S. military has been hiding evidence of a UFO landing in Roswell, New Mexico. It tried to dispel this by offering the usual bunkum that the UFO landing was actually a weather balloon, mistaken for a UFO. Physicist C.B. Moore of the New Mexico Institute of Science and Technology, an expert on observational balloon technology, skewered this twaddle. According to Prof. Moore,

"There were no such balloons in use back then, in 1947. The Skyhook series didn't even come onstream in New Mexico in 1947."

Moore ought to know as he was part of a high altitude balloon project based out of White Sands, NM. Adding:

"No balloon in use back then could have produced such a large debris field, over such a large area and torn up the ground as well."

This also douses the GAO report issued in 1995 that claimed the original AF weather balloon story was intended to conceal a more serious balloon -Skyhook - designed for military purposes to spy on the Soviets and provide early warning for a nuclear attack.

That one doesn't make the cut either given Prof. Moore noted "no balloon" back then could have produced the large debris field.

Strike two for the Debunk bot.

But hell, let's at least give 3 cheers to the bot for knocking out the clear conspiracy ideations (not 'theories'), such as that the Chinese created Covid 19 in a lab in Wuhan, and the falling of the Twin Towers was an "inside job."

Going back to the methodology: The conversations typically began with the chatbot summarizing the human's description of the conspiracy ideation. Then the human rated the degree to which he or she agreed with the summary on a scale from 0 to 100. From there, the chatbot set out to make the case that there was nothing fishy going on. To ensure the bot wasn't stretching the truth in order to be more persuasive, the researchers hired a professional fact-checker to evaluate 128 of the bot's claims about a variety of conspiracies. One was judged to be misleading, and the rest were true. Seems they missed the parts pertaining to the fake Oswald rifle and the Roswell UFO being a weather balloon, however.

The bot also turned up the charm. In one case, it praised a participant for ''critically examining historical events'' while reminding them that ''it's vital to distinguish between what could theoretically be possible and what is supported by evidence.''

Very solid advice which one hopes the bot and its creators follow more rigorously next time they take on real conspiracies. Lastly, one hopes that the researchers behind this useless project understand the enormous waste of energy as well as water, that's involved. See e.g.