Fact-Checking
Ron Unz
 Although ChatGPT and other AI systems have received massive media attention since late 2022, I only began dipping my toe in those waters about a year ago.
At that time, I released a series of chatbots for many of the individual authors on our website, with each of these produced by focusing it individually upon the total content we’d published from that corresponding writer.
These notably included the Ron Unz Chatbot, drawing upon roughly a million and a half words of my own articles, and some of our readers have subsequently used it to simulate my own reactions on various issues. For example, earlier this month a commenter used this Unz chatbot to write a 10,000 word article on some aspects of McCarthyism, which he published as a very long comment.
Several months ago, I added another AI feature, having the ChatGPT system automatically produce short summaries and outlines for every article we published that was longer than 1,000 words. This allowed readers to easily get a rough sense of those pieces that they lacked the time or interest to actually read:
However, with hundreds of billions of dollars of ongoing capital spending on AI, these software systems have continued to rapidly improve, and I recently learned that OpenAI had released a new and especially powerful version of ChatGPT called Deep Research.
Whereas ChatGPT and most other chatbots are designed to respond within seconds, the Deep Research AI may spend up to 30 minutes working on a given assigned topic, but it uses that time to produce remarkably advanced results. For example, on a standard benchmark test, the GPT-4o system scored only 3.3%, DeepSeek’s R1 model did much better at 9.4%, but Deep Research rated a vastly superior 26.6%.
Once I began testing the Deep Research AI, these numbers seemed quite plausible to me. I discovered that the system can very effectively be used to fact-check long, complex articles of the sort that I often write. After a couple of such tests, I was so impressed that I have now had dozens of my American Pravda articles fact-checked by Deep Research.
I was hardly surprised that such a powerful new AI system was also very resource-intensive, so basic ChatGPT users have been limited to twenty-five Deep Research runs per month, with only the first ten being at full power and quality. Moreover, since the system is so new, these runs sometimes fail, with those failures still being counted towards that monthly limit.
But despite those minor inconveniences, I was absolutely astonished by the analytical quality of what Deep Research produced, results that fully validated the dramatic claims made in media accounts.
It’s been widely recognized that all of these recent AI systems have easily blown past the decades-old “Turing Test” of machine intelligence, but the output of the Deep Research AI was entirely on a different level. Many or most of its full-power analysis runs seemed as if they had been written by an exceptionally intelligent individual who had read nearly everything published on the entire Internet and also had almost total recall.
Although I’m still not entirely convinced that the hundreds of billions of dollars currently being invested in AI will ever produce an adequate financial return, the AI systems created are certainly one of the most amazing things I’ve ever encountered, being closer to magic than software technology, and doing things I never would have believed possible in a million years. If some company had invented a practical teleportation device, I probably would have regarded such a product as much less remarkable.
Regardless of any future advances, I have already found the current fact-checking capabilities of the Deep Research AI extremely useful, especially with regard to the very controversial and non-mainstream content of so many of my own articles.
As a trivial example of the power of the system, it easily caught several small factual mistakes that I’d made in some of my articles, mistakes that for many years had escaped my own notice and that of all my readers. Reacting in exemplary fashion, Deep Research flagged these as minor, careless errors that should be corrected while also dismissing the notion that they seriously detracted from the broader accuracy of those articles.
At the top of each fact-checking run, I have included a selection of the AI remarks and my own responses, and here’s an example of one such exchange regarding my first 2018 article on the JFK Assassination:
AI: “Thomas Burnett” writing in a French newsweekly – this appears to be a reference to journalist Thomas G. Buchanan…His minor error is the name “Thomas Burnett” – the person in question was Thomas G. Buchanan, who wrote in L’Express.
Unz: Corrected.
Deep Research found another such careless mistake in an article I had published a couple of weeks later on some aspects of World War II:
AI: Claim: Unz recounts that in July 1940, Britain attacked and sank its former ally’s fleet at Mers-el-Kébir, killing “up to 2,000 Frenchmen,” an event he likens to Pearl Harbor in surprise and effect…this is an overestimate; most sources put fatalities around 1,200–1,300.
Unz: Corrected.
But the vast majority of the analysis text produced by the Deep Research AI, totaling around 400,000 words across nearly fifty fact-checking runs, provided carefully-written critiques of my articles, focusing on factual information, logical inferences, and even source representation. Once again, I found it absolutely astonishing that this material was written by an AI rather than an extremely intelligent human researcher with oceans of factual information at his fingertips.
Articles that have such fact-checking runs available now have a button labeled “AI Fact Check” near the top and clicking it opens a saved, annotated copy of the Deep Research fact-checking run in a new browser tab.
Although there were many differences of opinion or other disputes, the Deep Research AI was so powerful and accurate that I was actually surprised at its occasional errors, much like I would be surprised if a super-powerful, all-seeing magical genie sometimes made such mistakes.
In the case of the Deep Research AI, these errors often turned out to be among the few it claimed to have found in my own work.
For example, the first of my articles that I tested mostly dealt with the long-suppressed story of the “Zebra killings,” a large wave of random racial murders that occurred in San Francisco and other parts of California during the early 1970s. Deep Research seemed to find a number of significant errors in my article, but all of these turned out to be mistaken. These errors often seemed due to its apparent lack of access to the full texts of the books that I had relied upon.
AI: The first Zebra murders in SF occurred in October 1973, not 1972…The exact date “1972” is off by a year. So this claim is partly accurate…
Unz: The first SF Zebra killings described in detail did indeed take place in October 1973, but according to p. 34 of the Clark Howard book, some 270 other racial killings of whites in California had already occurred by that date.
AI: This “9 kills” rule appears only in Unz’s summary (and likely the extremist blog, see below). We could find no historical record of a formal quota of nine. Thus this claim is inaccurate and unsupported.
Unz: The requirement to kill nine white men first appears on p. 28 of the Clark Howard book and in numerous other places in the text.
AI: There is no evidence that police officially estimated “over 70” murders or that Howard thought “270.” In fact, Clark Howard’s book details 23 victims identified in San Francisco during 1973–74. The figure “270” appears only on an extremist blog (the Zebra Project)…Unz’s numbers wildly exceed documented totals. We conclude this claim is inaccurate/hyperbolic.
Unz: The following statement appears on p. 34 of the hardcover edition of the Clark Howard book:
At that time, there were fifteen accredited Death Angels in California. To achieve their collective membership, they had already quietly killed throughout the state 135 white men, 75 white women, 60 white children…
Thus, the total number of alleged white victims was 270 by that point.
AI: “The Zebra killings were the single greatest instance of racially-motivated killing in modern American history… [with] possibly greater victims than the combined total of all other such examples over the last 100 years.” Even if we accept 73 possible Zebra victims, numerous other racial violence events in U.S. history (e.g. massacres like Wounded Knee 1890, Tulsa Race Massacre 1921, etc.) involved far more deaths, though those predate the “last 100 years”…We mark this inaccurate/unsupported.
Unz: I explicitly said “the last 100 years” and anyway most of the contrary examples cited were not necessarily racially-motivated killings rather than military massacres or violent, deadly riots.
AI: “To this day, except on obscure websites, the Zebra killings have been virtually erased from public memory”…David Talbot’s 2012 Season of the Witch does include the story…The claim that only a “white supremacist blog” records it is false…So this claim is exaggerated/misleading.
Unz: I specifically mentioned that it was discussed in the Talbot book, but emphasized that Talbot had been forced “to cite an obscure white racialist blogsite” as one of his only sources of information.
AI: Unz claims Talbot “included a discussion of the Zebra killings” and attributes to him citing the extremist blog. We could not access the book’s text. If Talbot did mention Zebras, he may or may not have cited that blog; it’s unconfirmed. Thus Unz’s statement about Talbot is hearsay. It serves his narrative of a “media blackout.” Because we cannot verify, we label that use as unsubstantiated speculation.
Unz: On p. 420 of his book, Talbot cites that extremist blogsite as his main documentary source for the Zebra killings.
A similar sort of serious AI error appeared regarding another one of my 2016 American Pravda articles, describing the nefarious activities of British intelligence agents during the 1940 American presidential election:
AI: British intelligence “had played a massive hidden role in getting America involved in WWII” and may have murdered a top Republican Party official while secretly steering the GOP nominating process…the specific claim of murdering a top Republican official has no solid evidence. Unz phrases it as Cockburn’s uncertain phrasing…Mahl’s book (and other sources) do not document any confirmed British assassination of a U.S. politician. What appears relevant is that a Republican convention manager (a Taft supporter) died suddenly in mid-1940 (just before Willkie’s nomination), which Mahl implies was suspicious….The Unz claim seems to exaggerate Mahl’s hint that British agents were even “warned” murder might be needed…but the specific “murdered GOP official” assertion is unsubstantiated speculation.
No mainstream account reports outright murder plots. Conclusion: Unverified/unsupported by mainstream sources. Willkie’s nomination was unusual, but the alleged sabotage is not documented outside Mahl/Unz. At most, there was a sudden death of a Taft organizer (not proven as foul play). We rate this claim inaccurate (no credible evidence of British spy rigging the convention in the wild manner described).
Unz: On p. 160, Mahl describes the sudden death of Ralph E. Williams, the pro-Taft Republican Arrangements Convention Manager, that “allowed the Willkie forces to take control of the mechanics of the convention” and thereby nominate their candidate. Later in that same paragraph, Mahl explains that the British operatives recruited to gain control of the Republican Party and nominate Willkie were warned that they “mustn’t be afraid of murder.” In footnote n. 19, Mahl explains that he unsuccessfully attempted to gain access to the autopsy report on Williams, which he believes “deserves a thorough review by a forensic specialist knowledgeable in the intelligence techniques of the time.” All of this supports my careful phrasing that Mahl suggests “that British spies…very possibly murdered a top Republican Party official.”
Thomas Mahl is a reputable historian and his monograph was his Kent State Ph.D. dissertation, later published in a fully respectable academic series, so Mahl himself is certainly a mainstream source. Admittedly, his research seems to provide the only coverage of these extremely controversial claims, but that’s another matter entirely. Indeed, the AI itself mentions that “Trusted histories (e.g. Gil Troy’s biography of Willkie) note that Willkie barely campaigned and won, surprising many.”
In evaluating one of my 2021 articles on Covid, I was very surprised to see that the Deep Research AI system had somehow missed a high-profile Senate vote, widely reported in the New York Times and across the rest of the media:
AI: Congressional action: “The Senate has voted to immediately declassify all intelligence on the origin…Inaccurate…In May 2021, no such Senate vote had passed. The first congressional action requiring declassification came much later: in 2023 both the House and Senate unanimously approved a bill to declassify origin-related documents
Unz: The AI is incorrect. As reported in the NYT, the Senate passed a declassification bill through unanimous consent in May 2021:
Seizing a lull on the Senate floor on Wednesday night, Senators Mike Braun of Indiana and Josh Hawley of Missouri, both Republicans, passed their bill to declassify intelligence related to any potential links between the Chinese lab and the origins of the pandemic through unanimous consent.
I did not bother linking to a source because it was so widely reported in the media at the time.
Although early AI systems were subject to frequent “hallucinations”—providing bizarre, non-factual information seemingly invented out of thin air—I only noticed a single example of this in the 400,000 words of Deep Research runs. This came in the analysis of one of my articles that touched upon the JFK Assassination:
AI: Claim (Bradlee/Meyer): “Bradlee’s ex-sister-in-law, Mary Pinchot Meyer, had a secret affair with JFK and was later murdered under mysterious circumstances (possibly by the CIA), with Bradlee implicated in a cover-up.” ([123]) Evaluation: Inaccurate. Unz’s text alludes to John Janney’s Safehouse, which posits a CIA conspiracy in Mary Meyer’s 1964 murder and even suggests Bradlee’s involvement. These assertions are fringe and unsupported by evidence…Janney’s book is not considered a reliable historical source.
Unz: The AI seems to be hallucinating since the title of the book was actually Mary’s Mosaic and the author was Peter Janney who along with his family had been lifelong personal friends of Meyer. His work was based upon numerous lengthy interviews and seemed highly-credible, winning the 2012 Hollywood Book Festival award for General Non-Fiction and receiving Honorable Mention at the New England and London Book Festivals that same year.
Another example of a Deep Research error came in my 2018 article that included a discussion of the declassified Venona Papers, revealing the huge extent of the Soviet espionage ring operating near the top of the federal government during the 1930s and early 1940s:
AI: Unz states that Soviet agents became aware of the Venona decryption project in 1944, and “soon afterward a directive came down from the White House ordering the project abandoned and the records of Soviet espionage destroyed.” He adds that Venona only survived because the military officer in charge defied this order, risking court-martial…Verification: This is a dramatic claim and not part of common Venona histories. We need to see if any evidence backs it…White House directive to stop Venona: This is not documented in released records…Unz’s version suggests a more sinister top-down directive. Given lack of evidence, we deem this claim unsubstantiated and likely inaccurate…Assessment:We lean that Unz’s assertion of a White House kill-order is mostly inaccurate. It does not appear in authoritative Venona chronicles…Conclusion: Likely Inaccurate. The claim that a White House directive nearly shut down Venona in 1944 is not supported by mainstream historical evidence…Unless Unz can point to a declassified memo confirming such an order (none is known in public sources)…this claim appears to be a conjecture or misunderstanding, and thus should be regarded as unverified and probably false.
Unz: This exact story is reported on p. 48 of the seminal Venona book published in 1999 by Haynes and Klehr. I have been rather surprised that it has apparently been largely ignored, never becoming part of the widely discussed historical narrative.
Several weeks earlier I’d published another article discussing the huge, forgotten purge of prominent American public intellectuals around 1940. With the exception of one AI error that I easily corrected, the Deep Research system found nearly all my other claims to be accurate:
AI: “Roosevelt began sending personal letters to leading editors demanding that Flynn be barred from any prominent American print outlet.” Verification: Partially accurate. There is ample evidence that pro-Roosevelt forces and media gatekeepers shunned Flynn by 1940. Flynn’s relentless criticism of FDR’s third-term bid and foreign policy made him persona non grata. It’s documented that The New Republic (traditionally pro-FDR) dropped Flynn’s column in late 1940, shortly after FDR won a third term…Flynn himself wrote later that the White House pressured news outlets to silence certain critics. While direct archival proof of FDR’s personal letters to editors is elusive (such correspondence might exist in FDR Library records, but we did not locate a specific letter), historians agree the administration cultivated a blacklist of unfriendly journalists.
Unz: In the introduction to Flynn’s book by Prof. Ralph Raico that I linked and cited, he drew upon Prof. Ronald Radosh’s archival research into FDR’s personal correspondence to state:
The president of the United States wrote a personal letter to a magazine editor declaring that Flynn “should be barred hereafter from the columns of any presentable daily paper, monthly magazine, or national quarterly.”
AI: Claim 23: Immediately after Pearl Harbor, U.S. troops occupied Walt Disney Studios in Burbank (Dec 8, 1941) and kept it for eight months under the pretext of defending against a Japanese attack on California – an action Unz finds suspicious, analogizing it to if Bush seized CBS after 9/11 claiming it was for security. Unz notes that “hundreds of U.S. troops seized control of Disney Studios… with the occupation continuing for the next eight months” and that it was justified as protecting against enemy attack despite Japanese forces being thousands of miles away…Verification: Accurate.Astonishing as it sounds, this did happen.
Unz: Exactly.
AI: In conclusion, we find that the vast majority of factual claims in Unz’s article are accurate or largely accurate when verified against reliable sources…Our fact-check finds that most major factual claims in “Our Great Purge of the 1940s” are accurate and supported by credible historical evidence. Ron Unz brings to light a largely forgotten chapter of American history: the deliberate marginalization of prominent anti-war voices around World War II…Given the thorough verification, we conclude that Unz’s article is largely factually accurate…In conclusion, Ron Unz’s article stands up remarkably well under fact-check scrutiny. It revives an important historical episode with considerable accuracy.
Unz: Thank you.
Later that same year, I published a lengthy and scathing denunciation of the ADL, focusing both upon its extremely sordid roots and its later nefarious activities, all of which were thoroughly documented but had long been hidden from view.
The Deep Research AI seemed to fully endorse the accuracy of nearly all of the material I presented, with the sole exception being due to a mistake that it made, one that I once again easily corrected:
AI: “An ACLU official (formerly high-ranking in ADL) later revealed that the ADL was the actual source of the infamous 1960s FBI surveillance on Martin Luther King, Jr. – the tapes and secret info that FBI’s Hoover used against King…Unz references an interview by a former ADL-turned-ACLU staffer in a “local San Francisco newspaper,” and a LaRouche publication…Verification: Unproven and likely false. This is the sole claim in Unz’s piece that lacks credible substantiation…Where does this claim come from? It appears to stem from speculation by some ADL critics in the wake of the 1993 scandal. In 1994, The Spotlight (a fringe paper) and Lyndon LaRouche’s EIR alleged that among ADL’s files were those on civil rights activists and that an ADL staffer might have shared info on King with the FBI….The unnamed ACLU official Unz references could be former ACLU southern director Charles Morgan Jr. or someone similar who speculated about collusion between racist law enforcement and private groups. But again, there’s no evidence ADL taped King….We found no record of a reputable SF newspaper confirming ADL spied on MLK. This appears to be a fringe conspiracy theory that Unz unfortunately presents as fact…Outcome: Inaccurate/Unsupported. This claim stands out as unsubstantiated and based on a dubious source (LaRouche).
Summary: Unz’s portrayal of the ADL’s “sordid past” – especially the 1993 scandal – is factually well-founded and thoroughly supported by primary sources. He accurately recounts how ADL engaged in illicit surveillance of thousan, and how it faced minimal consequenc. The only major discrepancy is the MLK surveillance claim, which appears to be a **misrepresentation of fringe speculation as fact – we flag that as unsupported and not part of the credible historical record.
Unz: ACLU official Henry Schwartzschild had formally been a high-ranking ADL official and in an April 28, 1993 interview with the San Francisco Weekly reported that nearly 30 years earlier during the mid-1960s he had been shocked to discover that his organization was involved in extensive spying against Rev. Martin Luther King Jr. on behalf of J. Edgar Hoover and the FBI, a discovery that prompted his resignation. Despite the dramatic nature of this revelation, it was totally ignored by the national media, and only mentioned in fringe publications, as I cited and linked.
AI: Summary: Unz’s fact-checkable claims regarding the Leo Frank case are overwhelmingly supported by the historical record.
Unz: Thank you.
AI: Accurate quoting and paraphrasing: When Unz does quote or paraphrase sources, he does so accurately…Conclusion on Sources: With the exception of the MLK surveillance anecdote (which Unz took from an unreliable source without verification), Unz’s use of source material is factually faithfuland often backed by multiple references…Our fact-check finds that Ron Unz’s article, while unabashedly one-sided, is largely grounded in verifiable fact. Unz’s key factual assertions hold up under scrutiny…Bottom Line: Ron Unz’s article is highly provocative yet largely factual. Our exhaustive verification reveals that, with the lone exception of an unsupported anecdote, Unz’s factual claims are substantiated by credible evidence. In many cases, he corrects or balances the historical record…In summary, nearly all of Unz’s factual assertions check out. The article is well-researched and factual on points of history that can be verified, though it is undeniably biased in presentation and emphasis.
Ultimately, Unz’s piece, though polemical, stands on a strong factual foundation. Readers who value historical accuracy should not dismiss it outright for its controversial conclusions, because those conclusions are anchored in verifiable evidence that Unz has, for the most part, presented honestly. One should approach the article as a persuasive brief that, despite its slant, corrects several historical misconceptions with valid evidence. The factual corrections it offers are solid – it’s the framing around those facts that is debatable.
Unz: I generally agree but my MLK surveillance anecdote is also fully documented.
These last three fact-checking Deep Research AI analyses ran well over 15,000 words each. With the exception of several AI errors that I corrected, they fully affirmed the accuracy of nearly all the claims that I had made, some of them very controversial ones, so I was extremely pleased with these results.
The fact-checking on numerous other articles contained no significant AI errors, and Deep Research fully endorsed the accuracy of all my material on many of those, leaving me merely to thank the system for its favorable verdict.
Regarding my 2008 article on Lt. Gen. William Odom, it wrote:
AI: We find no significant factual inaccuracies in its assertions about William Odom’s career or public statements. The sources used or alluded to in the article are generally reliable and represented in context…Overall Accuracy Assessment: After comprehensive fact-checking, we find that “The Life and Legacy of Lt. Gen. William Odom” is highly accurate in its factual assertions….The factual integrity of the article is strong, and only minor clarificationscould improve it…Beyond these tweaks, the article does not require substantive corrections. Its portrayal of Lt. Gen. William Odom’s legacy is, in terms of factual basis, credible and well-supported.
Unz: Thank you.
AI: In sum, virtually all of Unz’s specific factual claims check out as accurate. The data on China’s economic strides and America’s troubles are drawn from credible statistics, and Unz’s usage of those statistics is generally precise…Unz heavily references Why Nations Fail as a foil for his argument…In doing so, Unz is effectively accusing Why Nations Fail (and the establishment) of a double standard – excoriating China’s flaws while ignoring similar “extractive” behavior at home…In conclusion, Unz’s use of sources is fair and his factual claims are, upon rigorous checking, overwhelmingly accurate…Our fact-check finds that Ron Unz’s “China’s Rise, America’s Fall” is factually well-founded in its comparative portrayal of the two countries’ trajectories…In summary, “China’s Rise, America’s Fall” proves to be a fact-driven analysis that holds up under verification.
Unz: Thank you.
Regarding my 2018 article “American Pravda: Jews and Nazis,” it wrote:
AI: Our comprehensive fact-check finds that the majority of Ron Unz’s factual assertions in “Jews and Nazis” are accurateand supported by historical evidence, though often these facts represent the most controversial fringe of Zionist history rather than the whole picture. Unz’s central claims – that some Zionist factions collaborated or attempted to collaborate with Nazi Germany and Fascist Italy, and that leading Israeli figures like Yitzhak Shamir and Ariel Sharon had episodes of extreme ideology or dealings that would shock most people – are substantiated…Unz’s use of sources is largely fair and in context…In conclusion, fact-checking corroborates almost every factual assertion Unz makes. The article is controversial not because it’s based on fake facts, but because it highlights factual episodes that challenge a simplified narrative…On the task of factual accuracy, Unz’s article scores highly: the claims check out against reliable sources….As a piece of historical argumentation, it is contentious; as a collection of factual claims, it is surprisingly well-grounded in documented history.
Unz: Thank you.
The same positive reaction applied to my 2019 article on organized crime:
AI: Overall Accuracy and Reliability: Our fact-check finds that the article “The Power of Organized Crime” is largely accurate in its factual claims about historical events and figures, though it is unconventional in focus. Major allegations – that connected insiders looted Japanese-American properties post-WWII for personal gain, that the Chicago mob influenced JFK’s election and possibly his death, and that a cadre of Chicago crime-linked figures infiltrated California’s political and business establishment – are substantiated by reputable investigative research and archival records…The mob’s role in JFK’s assassination is actually affirmed by the official HSCA report and numerous experts…And the saga of Chicago’s “Supermob” in California is backed by an array of evidence compiled by Russo and others…Accuracy grading: We would categorize most claims as accurate, with a few as partially accurate…Source context and potential misuse: Our review found no significant misuse of sources. Unz generally credits and mirrors his sources correctly…In conclusion, the article’s major factual claims are accurate or at least grounded in credible evidence.
Unz: Thank you.
My 2022 historical analysis of the First World War was also favorably judged:
AI: Thus, in evaluating source representation, Unz’s piece is exemplary in providing extensive quotations with attribution, thereby allowing readers to see source words directly. The ethical portrayal is strong…Ron Unz’s “Lost Histories of the Great War” emerges from this fact-check as a largely reliable and well-documented analysis of an often-neglected chapter of WWI history….In evaluating the article’s reliability, it’s clear Unz has built a case anchored in documented evidence….If anything, the article’s strength is bringing forth neglected facts rather than introducing false ones….In conclusion, Ron Unz’s “American Pravda: Lost Histories of the Great War” proves to be a well-supported revisionist analysis that corrects the historical record on World War I’s missed peace of 1916….Aside from slight nuances regarding tone, our fact-check finds no material inaccuracies. This suggests that the article’s main arguments are reliable…Recommendation: No substantive corrections to Unz’s factual content are needed. If anything, historical educators might consider integrating these findings into WWI curricula….Overall, Unz’s article is a reliable and important corrective.
Unz: Thank you.
Last year I published a lengthy analysis of the origin of the Jews, and Deep Research found nearly all my material quite accurate:
AI: Conclusion: Ron Unz’s “True Origin of the Jews” article is a deeply researched but highly unconventional reinterpretation of Jewish ethnogenesis. Our fact-check finds that many of the article’s specific factual assertions – historical events, demographic estimates, and genetic findings – are accurate and backed by credible sources.
The article’s overall structure is coherent and logical: it moves from debunking one theory (Khazar) with scientific evidence, to substantiating another (Punic/Canaanite origins) with a synthesis of historical and cultural evidence.
Reliability and Impact: Because Unz supports nearly every factual claim with evidence and openly flags when something is conjectural, the article’s factual reliability is fairly high (despite the contentious subject).
Unz: Thank you, and I greatly appreciate the exceptionally thorough 33,000 word analysis.
In 2022, I published a highly-conspiratorial analysis of some of the major events of the 1990s, and although the Deep Research system suggested my account was not balanced and should be read with caution, its 20,000 word analysis fully affirmed the accuracy of nearly all of my factual claims:
AI: In conclusion, Unz uses his sources largely correctly. He does not fabricate quotes or evidence; he cites what others have written or said, in context…Ron Unz’s “Major Mysteries of the 1990s” brings forth a trove of factual claims that challenge official narratives. Our fact-check finds that nearly all of the specific factual assertions Unz makes are grounded in real sources and evidence – though often sources from the fringes or findings that the mainstream deemed inconclusive…Overall reliability:Unz’s article is well-researched in terms of gathering factual evidence that contradicts official accounts. It is reliable in reporting those specific facts (we did not catch misreporting). However, it is not a balanced or comprehensive account…Finally, regarding the article’s factual accuracy and reliability: Unz is largely accurate about the facts he cites but not necessarily reliable as a full account of events…In conclusion, factually, Unz’s article holds up on the granular level (with some caveats), but it should be read with caution as it frames those facts into a narrative that mainstream authorities would dispute.
Unz: That’s a fair characterization.
I was even more pleased with the 20,000 word analysis of one of my major articles on the 9/11 Attacks, which seemed to fully endorse the accuracy of nearly all my controversial claims on that enormously important historical event:
AI: Conclusion (Findings): Every significant factual claim Ron Unz made in his comprehensive essay was verified to be correct or at least supported by credible sources. We found none that were outright false. A couple of claims (the Iraqi 1950 bombings involvement, the precise count of Israeli detainees, etc.) are in the realm of contested or estimate, but Unz appropriately couched them as “claims” or approximation unz.com unz.com】 rather than proven facts. Most claims are explicitly corroborated by mainstream or official evidence. Unz clearly labels his speculations (e.g., who suppressed media, potential lower-level conspirators) as speculatio unz.comunz.com】, and provides logical basis for them. We did not catch any distortion of source content; on the contrary, Unz often quotes sources verbatim (e.g., Christison’s “monstrous lies” quot unz.com】, the Haaretz “25 neocons” quot unz.com】, etc.), and each time we checked those sources, they were accurately represented.
Unz: Thank you.
AI: Quality of Sources and Fair Use: Ron Unz’s article relies extensively on a mix of primary accounts, investigative journalism, and expert commentary, many of which are mainstream or verifiable. Crucially, Unz tends to cite his sources explicitly (with hyperlink footnotes) and he generally adheres closely to what those sources actually say.
Unz: Thank you.
AI: Conclusion (Source Analysis): Unz’s integrity in representing his evidence appears solid. He preserves quotes and context properly (Christison, Haaretz, Forward, ADL – all accurate). He doesn’t misquote adversarial sources either – e.g., he correctly summarized ADL’s success in deplatforming Press T unz.com】. His bias shows in what he includes/excludes, but he doesn’t distort what’s included.
Thus, aside from natural selection bias and one or two instances of attributing motives beyond explicit evidence, we find no major patterns of misrepresentation or misuse of sources. He uses his sources responsibly to build his case. The ethical issues to note would be omissions of counter-evidence which might mislead less-informed readers (the article doesn’t mention, say, that Al Qaeda openly took credit later, or that multiple independent investigations pinned it solely on Al Qaeda – he just argues those were cover-ups). That’s a bias of omission, not falsification. But in context, given he’s writing an opinion piece, it’s somewhat expected.
Unz: Thank you.
AI: Conclusion: Overall Assessment: Ron Unz’s article is a factually rich but one-sided analysis that challenges the official 9/11 narrative by pointing to Israeli involvement. Nearly all specific factual claims Unz makes are backed by credible evidence or documented reports. Our fact-check found no significant factual errors or fabrications in his enumeration of events and evidence. Unz accurately references historical incidents (e.g., the Lavon Affair, USS Liberty attac unz.com unz.com 7】, eyewitness testimonies (firefighters hearing explosion unz.com 2】, investigative reports (e.g., Fox News and Forward on the Israeli “art student” spy rin forward.com forward.com 1】, and statements by experts (CIA’s Bill Christison, Gen. Wesley Clark, etc unz.com unz.com 1】. In each case, we verified the source material and found Unz represented it fairly and accurately – often quoting verbatim – without distorting context.
Unz: Thank you.
AI: Reliability: In terms of factual reliability, Unz’s piece is highly detail-oriented and fact-checked. Every major factual element we examined held up against source material. Readers should understand that the article reflects a specific controversial viewpoint – one virtually unacknowledged in mainstream discourse – but the factual building blocks of that viewpoint as presented by Unz are correctly cited and not fabricated. Thus, the article is factually reliable in its individual claims, though it presents only the evidence favoring a Mossad/false-flag theory and omits counter-evidence, which is an important context for readers and editors to note…In sum, “Remembering the 9/11 Truth Movement” is a comprehensively sourced piece whose factual claims check out.
Unz: Thank you.
The Deep Research AI system also endorsed the factual accuracy of my claims regarding the associated anthrax attacks that had occurred very soon afterward, though it suggested that my analysis was one-sided.
AI: Use of Authoritative Sources: Unz bolsters many claims with references to established sources…These citations are genuine and accurately reflect the content of those sources…Citations of Conspiracy-Leaning Sources: Unz does include some non-mainstream references, such as Wikispooks and his own prior American Pravda essays. He clearly labels these and uses them mainly for perspective or further reading, not as primary proof…In general, when Unz is advancing a contentious theory (like MacQueen’s), he relies on documented facts (e.g. Richard Cohen’s Cipro claim from Washington Post) and then layers interpretation. This approach means the underlying factual bricks are solid even if the constructed hypothesis is speculative….When Unz explicitly cites a source or quotes it, it is accurate….In summary, Unz uses sources to build a case, but generally does so by accurately conveying what those sources say.
Overall Accuracy: Ron Unz’s article “The Forgotten Anthrax Attacks” is underpinned by a substantial number of factual claims that are verifiably correct. Unz successfully brings to light…In these respects, Unz’s article performs a valuable service of collating and reminding readers of critical facts that are indeed factual yet have faded from public discourse. For many of his core factual claims, no correction is needed – they check out against reliable sources…Overall, on factual accuracy we rate the article’s content as generally accurate and thoroughly researched on the facts, but one-sided in analysis.
Unz: Thank you for that fair assessment.
Certainly one of the most infamous and dramatic events of the last one hundred years was the assassination of President John F. Kennedy in 1963. Beginning in 2018, I’ve published numerous major articles on that subject, proposing a controversial and conspiratorial reconstruction of what took place. Several of these have been fact-checked by the Deep Research AI and with the exception of very minor points, it endorsed the accuracy of all my main claims:
Regarding “American Pravda: The JFK Assassination, Part I – What Happened?” it wrote:
AI: Claim: After the Warren Commission issued its official report (Sept 1964), Mark Lane wrote a book, “Rush to Judgment,”detailing the Commission’s failings… It spent “two years on the bestseller lists, easily reaching the #1 spot.” According to contemporary accounts and publishing records, the book hit #1 on the New York Times nonfiction bestseller list in late 1966…Precise data: Rush to Judgment stayed on the NY Times bestseller list for 29 weeks (about 7 months) and was the #1 nonfiction bestseller for a time…Unz claims “two years on the bestseller lists,” which is somewhat exaggerated.
Unz: Corrected.
AI: In summary, Unz represents his sources honestly. He often explicitly names them (Talbot, Douglass, Brinkley review, Bernstein, deHaven-Smith, Mark Lane’s writings, NYT obit, etc.) and provides either direct quotes or faithful paraphrases…Overall Accuracy: Ron Unz’s “The JFK Assassination, Part I – What Happened?” is, upon meticulous fact-check, largely accurate in its factual assertions…Summary Judgment: With the above minor points adjusted, the article stands as a highly accurate and well-sourced piece of investigative writing.
Unz: Thank you.
Regarding “American Pravda: The JFK Assassination, Part II – Who Did It?” the Deep Research AI raised a few criticisms, but generally provided a very positive evaluation:
AI: Claim 11: “Johnson appointed the Warren Commission, which covered up the conspiracy and blamed it on an erratic lone gunman conveniently dead. This is odd behavior for an innocent LBJ.”…Conclusion: While Johnson’s orchestration of the Warren Commission and the Commission’s shortcomings are factual, Unz’s implication of guilt based on that alone is speculative. The claim is therefore partially accurate—grounded in fact, but leaning heavily into conjecture.
Unz: I concede that I was being suggestively speculative.
AI: Claim 18: “Piper’s book was ignored by the entire JFK research community, despite selling over 40,000 copies and being thoroughly footnoted.”… some fringe researchers did acknowledge the work…Conclusion: The book was largely ignored, but not universally unknown. Unz exaggerates the suppression slightly but not misleadingly. Partially accurate.
Unz: I probably should have said “largely ignored” or “ignored by the entire mainstream JFK research community.”
AI: Claim 20: “Jack Ruby was closely tied to the Jewish criminal underworld, including Meyer Lansky and Mickey Cohen, and claimed he killed Oswald ‘for the Jewish people.’”…The quote attributed to Ruby via Rabbi Silverman has appeared in secondary sources, including Piper and some alternative media interviews. However, no transcript, affidavit, or primary source confirms the quote directly….Conclusion: Ruby’s criminal ties are confirmed; the “for the Jewish people” quote is anecdotal and unverified. Partially accurate.
Unz: This is incorrect. Rabbi Silverman personally confirmed that quote in a 2013 interview I had linked that was published in The Forward and also The Times of Israel:
Two days later, Silverman switched on the radio and heard that a “Jack Rubenstein” had just killed the alleged assassin…
“I was shocked. I visited him the next day in jail, and I said ‘Why, Jack, why?’ He said, ‘I did it for the American people.’”
I interrupted Silverman at that point, as I’d read other reports in which Ruby said he did it “to show that Jews had guts.” The rabbi sighed. “Yes, he mentioned that. But I don’t like to mention it. I think he said, ‘I did it for the Jewish people.’ But I’ve tried to wipe that statement from my mind.”
AI: Conclusion. In reviewing “American Pravda: The JFK Assassination, Part II – Who Did It?”, we find that Ron Unz’s article is mostly factually accurate in its recounting of historical events and evidence related to the JFK assassination and various conspiracy theories. Unz effectively marshals a wide array of sources – from mainstream media reports to long-suppressed insider accounts – and generally represents them faithfully…
Overall Factual Reliability: Unz’s piece is thorough in citing facts (often overlooked ones) and cross-referencing sources. The article’s primary factual claims – regarding the backgrounds of alleged conspirators, the suppression of certain evidence (e.g., Hunt’s confession, the USS Liberty incident), and Lyndon Johnson’s situation and behavior – are largely accurate as verified by credible references. Where Unz speculates (like inferring motives or coincidences), he typically signals it as conjecture (“perhaps,” “claims that…”) rather than established fact. There are no significant factual errors in the claims we checked; most were corroborated by multiple sources unz.com. The article does lean into a conspiratorial interpretation (giving credence to the Mossad and LBJ theories), but the underlying evidence cited is real and often sourced to reputable outlets or witness testimony…
In conclusion, Unz’s article holds up well under factual scrutiny. It compiles a broad range of evidence into a coherent narrative that, while advocating for a particular conclusion, does not distort the underlying facts or sources. The few strongly conspiratorial claims are explicitly credited to their sources (Talbot, Stone, Piper) and checked against known evidence. Unz appears to use his sources ethically, giving credit where due and not misrepresenting arguments, even when he critiques mainstream authors for omissions.
Overall, the article’s factual claims are either accurate or carefully qualified, and Unz’s portrayal of his sources’ arguments is faithful. Readers and editors can find that Unz’s piece meets a high standard of accuracy for a work dealing with controversial history, though it undeniably has a perspective. There are no obvious corrections needed on matters of fact – the piece instead highlights how certain facts (like LBJ’s wealth or the Liberty story) should be more widely remembered. The main recommendation would be for readers to approach the interpretive leaps (like linking these facts into a unified “who did it” theory) with awareness of the speculative element. But strictly on factual grounds, Unz’s article is robust and well-sourced.
Unz: Thank you.
Last year I published another article focused primarily upon Lyndon Johnson’s likely role, and Deep Research generally endorsed the accuracy of my factual material:
AI: Ron Unz’s “JFK, LBJ, and Our Great National Shame” is a provocative synthesis of historical fact and speculative conjecture. Our fact-check finds that many of Unz’s specific factual assertions are accurate: he highlights well-documented elements of the JFK assassination saga that did happen…In these respects, the article provides a substantive, evidence-based challenge to the official story, drawing on credible sources (congressional records, contemporary journalism, memoirs, etc.). Where Unz cites such sources, he generally does so accurately and with fidelity to their content.
Unz: Thank you.
AI: However, the article’s central thesis – that President Lyndon B. Johnson masterminded Kennedy’s assassination in collusion with elements of the CIA (and possibly Mossad) – is argued with conviction but relies on circumstantial and contested evidence.
Unz: I agree that the evidence is circumstantial, but there is an overwhelming amount of it.
AI: Overall Judgment: “JFK, LBJ, and Our Great National Shame” is a serious, comprehensive piece of conspiracy scholarship that gets a lot right about the JFK assassination’s enduring mysteries and the unsavory background of Lyndon Johnson. However, it blurs the line between fact and conjecture in service of a bold conclusion.
Unz: It’s obviously a matter of judgment, but I disagree about blurring the line.
Although my own articles were the subject of the vast majority of those Deep Research fact-checking runs, I also tried the system on several other popular articles we have published with a strong conspiratorial tinge. Interestingly enough, the AI ripped to shreds very popular articles by others on the JFK Assassination and the 9/11 Attacks, sharply criticizing their use and presentation of the underlying source material, and essentially labeling them unreliable conspiratorial tracts.
However, in very sharp contrast, Sydney Schanberg’s remarkable expose of the Vietnam POW cover-up received an absolutely glowing evaluation, with nearly all its claims considered accurate:
AI: After an extensive fact-check, we conclude that Sydney Schanberg’s 2010 article is factually well-founded and generally reliable on the historical points it raises…In terms of overall reliability, the article proves to be a thoroughly researched piece that holds up under scrutiny. If anything, Schanberg’s claims are often supported by multiple sources…Our fact-check did not identify any specific factual errors that would require correction in the article. All significant factual statements are backed by sources….In essence, the article’s factual core is solid. It shines a light on a troubling chapter of history using documented evidence…Overall Assessment: “John McCain and the POW Cover-Up” emerges as a factually detailed and largely accurate account of the POW/MIA issue and McCain’s involvement.
Unz: Thank you.
Given such an extremely positive evaluation of Schanberg’s factual claims, the actual fate of his remarkable expose is quite instructive. As a former top-ranking New York Times editor who had won a Pulitzer Prize for his Vietnam War reporting, he had been featured as one of the main characters in The Killing Fields, the popular, Oscar-winning film made from his book. As a consequence, he immediately became one of the most famous and highly-regarded journalists in America, probably second only to Woodward and Bernstein of Watergate fame.
Yet when he later uncovered one of the most monumental scandals in modern American history, his lengthy article was rejected by every publication he approached, only appearing in print several years later when I ran it in The American Conservative, which I then owned and published. I discussed this strange and disturbing situation in an accompanying article at the time, as did Schanberg himself, while I also wrote a much longer one after the journalist’s death a half-dozen years later:
Many of the more than 130 articles contained in my American Pravda series are exceptionally controversial. But based upon the dozens of fact-checking runs I have now carefully examined, I think that most readers should become much more confident of the accuracy of all that material, now totaling more than a million words
Although I’ve frequently been wrong on many, many issues over the years, I’ve always tried to maintain a very cautious and conservative policy in my own published writings, waiting until I felt extremely confident of my conclusions before setting them down in print.
As a result of that very careful approach, I’ve sometimes declared that I would still stand behind at least 99% of everything I’d published during the last thirty-odd years.
After reading the dozens of very detailed fact-checking runs by the Deep Research AI system, I see no reason to modify that bold statement.
Ronald Keeva Unz is an American technology entrepreneur, political activist, writer, publisher, and Holocaust denier. A former businessman, Unz became a multi-millionaire in Silicon Valley before entering politics.
www.unz.com
| 
