A six-person fact-checking team at India Today spends its days sorting through false and misleading images: AI-generated archival photos of politicians, flood footage taken out of context and deepfakes of prominent figures.

Now the team is testing Backstory, an experimental tool from Google DeepMind that brings several parts of that verification process into one place, according to Nieman Lab’s Andrew Deck.

A reporter can upload an image and ask Backstory a question. The tool then decides which checks to run. It can make a SynthID tool call to look for watermarks embedded in images created with Nano Banana, Google’s image-generation model. It can also check content credentials tied to the Coalition for Content Provenance and Authenticity standard and trace where an image has appeared online.

The result is an AI-generated report with citations and a log of the steps Backstory took.

The potential time savings are a major attraction. Mike Caulfield, the digital literacy researcher who created the SIFT method, told Nieman Lab that Backstory can shrink one initial verification process that might take 50 minutes to about three.

Bal Krishna, who leads India Today’s fact-checking team, described the benefit another way: “It is doing all the work you might have done with five different tools, five different logins, in the same place.”

Backstory is free for now but available only through Google’s Trusted Testers Program. DeepMind has given access to thousands of journalists, open-source intelligence researchers, librarians and researchers.

That newsroom input has not always been part of the development process for authentication tools. A July report from Princeton University’s Center for Information Technology Policy found that many such tools are built without enough consultation with journalists. Some produce confidence scores about whether an image is synthetic while offering users little clarity about how those scores were reached or how they should be explained.

Backstory also makes mistakes.

When Deck uploaded his headshot, the tool confused it with a photo of another journalist and generated a report about the wrong image. Caulfield described the mistake as “conflation,” a problem that can occur when large language models encounter visually similar images.

There are other limits. Backstory currently works only with still images, and its context-tracing feature can find only material indexed by search engines, leaving gaps on platforms such as TikTok and closed messaging services.

India Today treats the tool as a starting point rather than a source. Krishna said reporters manually confirm information they get from Backstory, and the organization does not cite its AI-generated reports directly in published fact-checks.

The tool also comes from a company producing the kinds of synthetic images fact-checkers increasingly encounter. Google added Nano Banana image generation to Google Earth in July, then said it would roll back the feature while developing stronger guardrails after users shared misleading generated satellite-style images.

Zoe Darmé, a DeepMind product manager, told Nieman Lab that Backstory was not developed as a direct response to criticism of Google’s image tools.

For Krishna, the appeal is simpler. Asked whether Backstory has sped up his team’s daily work, he said: “There is no doubt about it.”

The post Google’s Backstory aims to speed up image verification for fact-checkers appeared first on The Media Copilot.

Anthropic is offering new details about how it plans to watermark text written by Claude — including what happens when someone edits the output.

The company uses a simple example to explain the technology. If Claude can choose between words such as “overcast” and “grey” to describe the weather, it can make those small choices in a pattern. A reader wouldn’t notice. But someone with the right detection tool could identify the pattern as a watermark.

Anthropic described the system in a Friday TechCrunch blog post answering questions about the plan, which it announced earlier in the week to comply with the European Union’s AI transparency rules. We covered the original announcement here.

The company plans to use SynthID Text, a watermarking technology developed by Google DeepMind. Anthropic also says it will release an API to detect the watermark.

The system has limits, particularly once people start changing Claude’s words.

“Light editing probably won’t remove the watermark completely,” Anthropic said. But a complete rewrite in which every word is replaced would remove it. Anthropic said in such a case, it becomes harder to describe the finished product as AI-generated.

Whether a watermark can be detected also depends on how much Claude actually wrote. If someone gives Claude a human-written document and asks it to make only minor edits, Anthropic said that leaves “very little (if anything) for the watermark to attach to.”

Code presents a similar problem. Code gives a model fewer stylistic choices than prose. Anthropic said it can still watermark choices, including some comments and names within code, but the technology should have “a negligible effect on the actual code produced.”

The company is also drawing a distinction between watermarking and statistical AI detectors.

Those tools look for patterns associated with AI writing and estimate whether a model produced a piece of text. A watermark works differently, using a pattern deliberately inserted when the text was generated.

An AI detector makes a statistical judgment. A watermark provides evidence that a particular system marked the text when it was created.

Not everyone is happy about the change. TechCrunch reported that dozens of Claude users on X said they had canceled subscriptions over the watermarking plan, citing Business Insider.

For news organizations, the technology could make questions about AI use easier to investigate.

An editor could check whether copy submitted by a freelancer carries Claude’s watermark. But competitors, critics and others with access to Anthropic’s detection system could check published work as well.

That makes newsroom policies around AI use more consequential. Publishers may need clear rules about when it is permitted, how much AI-generated material can appear in published work and when that use should be disclosed.

Claude won’t necessarily be the only model leaving marks behind. Anthropic says other major AI developers that signed the same European code of practice are also working on watermarking systems.

What remains unclear is whether those systems will work together. Publishers find themselves checking text against a different system for every AI company.

The post Anthropic says Claude’s text watermark survives light editing but not rewrites appeared first on The Media Copilot.

A PR agency representative offered guaranteed placement on a website impersonating Forbes for $10,000, according to a recorded sales call obtained by Press Gazette. Forbes later confirmed that forbesla.com has no connection to the company and said it is taking action against it and another site posing as a Forbes publication.

The two fake Forbes sites publish dozens of stories a day under bylines with little or no apparent presence elsewhere online, Press Gazette first reported. Reporters Nathan Pine and Liam Redmond appear to have no social media profiles. AI detection tool Pangram classified one of Redmond’s articles as entirely AI-generated.

Press Gazette also reported that Identifai flagged the byline photo of Harriet Caldwell, described by Forbes Liechtenstein as an experienced news editor, as AI-generated. Her articles were also flagged as likely to have been generated by AI.

The sales pitch extended beyond the fake Forbes sites. Baden Bower, a public relations agency that promises clients guaranteed media coverage, was the company whose representatives offered the placements documented by Press Gazette. One representative, Zac, offered placement in the Los Angeles Times, said The Guardian and The New York Times were the only publications where he could not secure coverage, and quoted $9,900 for a Bloomberg TV interview with an Emmy-winning journalist. He also suggested paid articles could help applicants seeking U.S. skilled worker visas.

A second Baden Bower representative, Camilo, said he was unaware of Forbes LA and that the agency works only with licensed regional Forbes editions. He quoted $25,000 for Forbes US, $2,000 for Forbes Georgia and $100,000 for The Wall Street Journal.

Asked whether paid placements are labeled as promotional, Camilo said, “Some are, some are not.” He also claimed Baden Bower has commercial relationships with more than 1,980 publications, including Press Gazette. Press Gazette said it has never published a sponsored article from the company.

Much of the content on the fake sites promotes companies, including Baden Bower itself. One Forbes Liechtenstein article, which Pangram classified as entirely AI-generated, projects that the agency will reach 7,600 clients and 90,000 published stories by the end of 2026 and quotes company leader AJ Ignacio touting its commercial results.

Baden Bower then cites that article on its own website as evidence of its standing, alongside pieces from an AI-generated Philippines edition of Fast Company and an apparent fake German edition of Vanity Fair.

The sites are not new. The Internet Archive shows Forbes LA online since January 2025 and Forbes Liechtenstein since March 2025. The London office listed by Baden Bower has a single contact number, a mobile number that appears to be disconnected.

For publishers, the risk is to the masthead itself. Fake sites can cheaply trade on the authority of established news brands, turning their names and reputations into promotional inventory. The model echoes fake AI news networks built to game search results and fabricated AI expert profiles reaching real reporters, but extends the problem into the PR industry.

Sarah Waddington, CEO of the Public Relations and Communications Association, said the practice violates industry standards on transparency, honesty and earned credibility. Former BBC journalist Omar Hamdi, who first noticed founders sharing fake Forbes coverage on LinkedIn, said the promotional tone of the articles raised his suspicions.

For newsrooms, the simplest defense may be basic verification: check that the publication behind a citation actually exists before trusting it.

The post Fake Forbes sites sold guaranteed coverage for $10,000, PR pitch shows appeared first on The Media Copilot.

Runtime Wire has published 1,627 articles since launching in May, after scanning 71,796 potential stories and publishing just 2.3% of them, according to a profile in Talking Biz News.

The site covers the AI economy, from funding rounds and model launches to developer tools, infrastructure, executive moves and court filings. Its stories have drawn 205,000 pageviews.

And the newsroom has exactly one person on staff.

“I’m the founder and the only human,” said Ryan Merket. “The newsroom itself is software, a pipeline of AI systems that scans sources, researches, writes, edits, fact-checks, and produces the video and audio around the clock.”

Merket describes his job as part editor-in-chief, part engineer. He sets editorial standards, reviews coverage each day and builds the systems that handle the volume. He previously served as CTO at Microsoft for Startups and worked at Reddit and Amazon Web Services.

The pipeline monitors about 50 sources, including vendor blogs, changelogs, research feeds, social media and tips. Off-topic material, such as sports or crypto price chatter, is filtered out before scoring.

The remaining stories are evaluated by an AI curator against a written editorial standard: Is the story new? Does it matter to people building with AI? Does it come from a live primary source? Has Runtime Wire already covered the underlying event?

Two additional layers check for duplicate coverage and re-reports.

What happens next shows where the automation actually sits. A research step fetches and reads primary sources, including attached images, which are transcribed so image-only claims can be checked. Merket says the research brief is anchored to the current date, helping prevent old news from being presented as breaking.

A writing model with live web search drafts the headline, summary and story. An editor model then reviews the piece for depth and accuracy and can send it back for more research. A separate fact-checking pass tests specific claims against the live web.

For stories Merket assigns himself, he serves as the editor and reviews them before publication. In the automated lane, no human reviews the copy before it goes live.

Runtime Wire has not generated revenue yet. Merket says its monetization infrastructure is already live, including sponsored placements in code editors such as VS Code and Claude Code, a jobs board and site sponsorships.

For now, he is running ads against house inventory to tune pacing, quality scoring and bot filtering before selling placements more broadly. The strategy is B2B-first, targeting developer-tools advertisers while keeping the site free for readers.

For newsrooms, the more revealing number may be the 2.3% publish rate. Runtime Wire is not betting on volume for its own sake. Its pitch is that most of what it scans never gets published.

That sets it apart from the AI content farms that have drawn scrutiny from watchdogs such as NewsGuard, which has tracked hundreds of unreliable AI-generated news sites. Whether a one-person newsroom can maintain that level of editorial filtering as it scales remains the bigger question.

The post Runtime Wire runs an AI newsroom with one human and 1,627 published stories appeared first on The Media Copilot.

Reuters treats AI as a research assistant that must show its work. The BBC requires a human sign-off before any AI-assisted story runs. The Guardian will only let staff use it once a senior editor has approved the specific task. Three of journalism’s most prominent outlets, three distinct answers to the same question — and together they show there is no single blueprint for the AI-powered newsroom.

A new comparative study published by Nieman Lab examined how Reuters, the BBC and The Guardian have approached AI deployment, finding that each outlet has built its own framework around editorial priorities, governance and audience trust rather than embracing a uniform set of tools. The research analyzed public AI policies, editorial guidelines, trial reports and industry presentations released through 2025.

The study suggests AI adoption in journalism is increasingly less about whether to use the technology and more about how news organizations establish controls over its use.

Reuters has positioned AI as a reporting assistant that speeds journalism without replacing editorial judgment. Its in-house tools include Lynx Insight, which analyzes financial data in real time, and Fact Genie, which helps reporters process corporate press releases. Journalists must independently verify AI-generated claims before publication under its standards. Market reporting also provides unusually fast error-correction signals, since traders spot bad information quickly — a less forgiving model than political reporting or investigations, where a factual error can sit unnoticed far longer.

Reuters Editor-in-Chief Alessandra Galloni has described AI as a “force multiplier” that helps journalists analyze large document collections and identify patterns more efficiently, with governance checkpoints built in at each stage to preserve editorial accountability.

The BBC has taken one of the industry’s most cautious approaches, prioritizing transparency over rapid deployment. It permits AI for specific tasks such as At a Glance, which summarizes long articles, and BBC Style Assist, which reformats local stories in house style — a task the broadcaster says can otherwise take about 30 minutes. No AI-assisted material runs without an editor’s sign-off first.

That caution is evidence-led: in October 2025, the BBC and the European Broadcasting Union published research showing AI assistants misrepresented news content 45% of the time through faulty sourcing, fabricated details and outdated information. The broadcaster later introduced AI disclosure labels at the top of stories.

The Guardian revised its guidance in March 2026 to permit limited use of generative AI for tasks such as generating image alt text, analyzing parliamentary documents and transcribing audio, provided the tools meet editorial standards and a senior editor approves where required.

The policy reflects the outlet’s broader philosophy that AI should support, not replace, journalists. “Our generative AI framework is designed to support our journalists’ expertise, never replace it,” the outlet said in a statement. “Our journalists are always accountable for the journalism they create, and use of these tools requires absolute rigor and responsibility.”

The Guardian’s reader-funded trust ownership model has also shaped its strategy, allowing the publisher to prioritize editorial independence over commercial pressure even as the newsroom gradually expands its use of the technology.

The findings illustrate an industry shift away from viewing generative AI primarily as a content-generation tool. Leading publishers are increasingly deploying it behind the scenes — for research, document processing, transcription, translation and newsroom efficiency — while keeping journalists responsible for editorial decisions.

The divergent strategies also reflect broader debates across the news industry over governance, transparency and the economics of AI. As publishers negotiate licensing agreements with technology companies while pursuing litigation over unauthorized use of copyrighted reporting, many are simultaneously developing internal policies governing how AI can be used inside their own newsrooms.

Nieman Lab’s analysis concludes that newsroom AI adoption is evolving into an organizational question rather than simply a technological one. Success, the report argues, depends not only on selecting AI tools but on establishing editorial policies, accountability mechanisms and workflows that preserve journalistic standards as AI becomes part of everyday reporting.

The post Reuters, BBC and Guardian chart distinct newsroom paths for using AI tools appeared first on The Media Copilot.

James Allen of Billpin was quoted as a finance expert by outlets including CBS News, the New York Post and Business Insider. Qwoted later barred his account after flagging repeated AI-generated submissions and finding that he had failed its verification process, according to Press Gazette’s investigation.

The case is a stark warning about the source pipeline behind everyday newsroom copy. Qwoted, a U.S. platform that connects journalists with sources, says it has banned hundreds of agencies, executives and experts for conduct ranging from fabricated identities to AI-generated pitches and false claims of expertise.

Billpin’s website is now blank, and Allen’s LinkedIn profile has disappeared, Press Gazette reported. The publication said Allen did not respond to requests to verify his identity. Press Gazette concluded the operation appeared designed to win media mentions that could boost search rankings, or backlinks.

A quote in a respected publication can generate a link that helps an otherwise obscure website gain visibility in search results. That creates an incentive to manufacture expert identities that appear credible enough to earn media coverage.

Qwoted CEO Dan Simon said generative AI has dramatically accelerated an already difficult problem. “AI has just poured petrol on the disinformation fire,” he told Press Gazette.

The platform says it now handles more than 15 violations a day and blocks about 50 prospective clients from registering each day. Shelby Bridges, Qwoted’s head of user experience, said the volume of low-quality pitches began rising rapidly after ChatGPT’s public release in November 2022.

Simon described a source ecosystem that increasingly blurs the line between legitimate expertise and fabrication. “You’ve got real agencies representing fake experts with fake expertise. You’ve got real agencies with fake experts and real agencies with real experts with fake expertise, and fake agencies with fake experts with fake expertise.”

One recurring pattern is a purported expert responding to queries on subjects with no obvious connection to their background. Press Gazette identified examples of sources offering commentary on finance, cleaning, human resources, pet sitting and mental health.

Qwoted combines human review with signals such as unusually high pitching volumes. It also uses Pangram, an AI-content detection service, to flag potential abuse. But AI detection is only one signal. A model score cannot establish that a person is fictional, that a quote is false or that an expert lacks relevant experience. Qwoted’s reported checks therefore matter more than a detector’s output alone.

Simon said some reporters have told him they do not object to real experts using ChatGPT to polish their responses. The bigger risk comes when AI helps conceal a fabricated identity or invented credentials, allowing a bogus source to slip into print before anyone questions its legitimacy. Platforms can suspend accounts after the fact, but the final responsibility still rests with newsrooms.

The post Qwoted bans AI-generated expert profiles as PR spam floods reporters’ inboxes appeared first on The Media Copilot.

Substack readers can now run a check on any post longer than 100 words to see how much of it an AI model likely wrote. The company rolled out the feature on July 21 in partnership with the detection firm Pangram, with the scan available on web and iOS and Android support promised later.

The tool works only on content published starting on July 21. Alongside it, Substack is giving writers a dedicated space to state whether and how they used AI in a given piece. The idea, per the company, is to set reader expectations rather than police anyone.

No AI detector is reliable enough to call a piece machine-written with certainty. Pangram uses a deep learning classifier trained on around a million documents rather than the older perplexity-and-burstiness heuristics, and it reports strong internal accuracy numbers. But The Atlantic examined how well Pangram actually performs and found the picture more complicated than the marketing suggests. False positives are a known problem across the category, and lightly edited or paraphrased AI text is far harder to catch than raw model output.

Substack did acknowledge those limits in its announcement post. It also hinted at further features tied to AI content and reader preferences without saying what they are.

Its acknowlegdment matters because a wrong flag carries real cost. A human writer who gets tagged as an AI user has little recourse in the moment, and the accusation sticks. Non-native English writers have historically drawn more false positives from detection systems, which raises fairness questions for a platform with a global base of newsletter writers. Pairing the automated score with a voluntary disclosure field is Substack’s hedge against that, letting writers put their own account on the record next to the machine’s.

For newsrooms and independent publishers, the move is a preview of how disclosure norms may harden across platforms. If readers start expecting an AI provenance label on every post, editors will face pressure to document their own use of the technology and to decide where the line sits between an AI writing assistant and AI ghostwriting. Publishers who have built AI into their workflows, a shift The Media Copilot has tracked closely, will want a clear internal policy before a third-party score answers the question for them.

The real test isn’t whether the detector works. It’s whether readers trust a score that Substack itself acknowledges can be wrong. The company is betting users will treat AI detection as one signal among many, not a definitive verdict.

The post Substack adds AI detection to its posts through a new Pangram partnership appeared first on The Media Copilot.

In 2023, a company called Alabama Community News LLC supposedly spent $3.2 million to buy 47 weekly newspapers across the state. The corporate owners fired the local staff, replaced them with an artificial intelligence system that scraped high school sports scores, and promptly drove the entire network into bankruptcy. The story even named a specific 26-year-old campaign staffer who generated 70 percent of the copy.

None of it actually happened. The entire 1,900-word saga was a fabrication published by a site called The Editorial, according to an investigation by Nieman Lab. The targeted newspapers, including the Shelby County Reporter and the Centreville Press, are still printing. The angry local advertisers quoted in the piece do not exist. The story falsely claimed the roll-up was funded by 1819 News, a real conservative outlet in the state, adding a layer of plausibility to the hoax.

The fake story gained traction among journalists on social media platforms like Bluesky before the operators pulled it down. They replaced the page with a sterile retraction notice citing “fact-verification concerns.” But the Alabama hoax was not an isolated incident.

The Editorial has built a bizarre subgenre of AI-generated obituaries for real American newspapers. The site previously published fabricated stories detailing the collapse of the Chattanooga Times Free Press, the Kenosha News, and the Macon Telegraph. The nonexistent reporters credited with these stories sport fake resumes claiming past stints at ProPublica and Reuters.

The motive behind the site remains murky. Domain registration and payment records point to a Finnish technology company called Nordiso Group, which develops AI study apps. Yet the site’s political sections suggest a different angle. The Editorial publishes a high volume of geopolitical content focused on Taiwan and the South China Sea, heavily pushing narratives that highlight Chinese military dominance.

These geopolitical stories share obvious synthetic fingerprints. Nearly every piece opens with a variation of the exact same scene: a nondescript, windowless conference room where a secret document slides across a table. This repetitive structure aligns with tactics tracked by groups like the Stanford Internet Observatory, which monitors state-sponsored disinformation campaigns. It also highlights how cheap synthetic media allows operators to flood niche topics, a trend we track closely at The Media Copilot.

For publishers, this represents a strange new vector of reputational risk. Newsrooms are used to fighting disinformation about elections or public health, but now they must monitor for synthetic hoaxes about their own business operations. A fake story about a newspaper shutting down or firing its staff can spook actual advertisers and confuse real subscribers before the publisher even realizes the rumor exists.

The barrier to generating convincing local news copy is gone. Operators no longer need to understand the nuances of a community to write a plausible story about it. They only need a prompt and a target, leaving local editors to clean up the mess when the synthetic fallout hits their own backyards.

The post AI fake news network invents the collapse of 47 local Alabama newspapers appeared first on The Media Copilot.

NewsGuard, a company best known for rating the reliability of online news sources, on Tuesday launched NewsGuard AI, a chatbot that draws exclusively from a database of journalist-vetted stories instead of the open web.

The launch comes as concerns persist over the accuracy of AI-generated responses. NewsGuard said a yearlong audit of leading AI models found they repeated false or misleading claims on controversial news topics 35% of the time. The company argues that limited responses to vetted sources can help reduce the spread of misinformation through AI systems. 

NewsGuard AI attributes information directly to the publishers whose reporting is used in its responses, unlike other chatbots like ChatGPT, Claude, Gemini, or Perplexity.

Participanting publishers include The Atlantic and other regional newspapers, opinion journals, and public media organizations. Readers, subscribers and members of some participating outlets will receive a free trial of NewsGuard AI followed by an offer for 33% off the chatbot’s standard $6 monthly subscription. 

The company also says it will share revenue with participating publishers through a 50-50 revenue-sharing model and affiliate-style subscription referrals, though it has not publicly disclosed the formula used to calculate payouts.

NewsGuard says its journalists have reviewed more than 36,000 sources since 2018, including newspapers, magazines, opinion publications, local news outlets, independent newsletters, government websites, think tanks, hospitals and research universities. Of these, roughly 12,000 have been rated reliable and are eligible to be cited by NewsGuard AI. 

The new service enters a rapidly evolving market in which publishers are negotiating licensing agreements with AI companies while also challenging the unauthorized use of their reporting. Media organizations have struck content deals with companies including OpenAI, Amazon and Meta, even as lawsuits and public disputes over AI scraping continue across the industry.

Chris Richmond, CEO of the fact-checking website Snopes, said the arrangement addressed concerns his organization has had with other AI products.

“Snopes has restricted most AI chatbots from scraping our content,” Richmond said. “But we’re happy to partner with NewsGuard on a model that does this the right way.” 

In addition to drawing from vetted sources, NewsGuard AI says it incorporates 41 editorial safeguards. These include access to NewsGuard’s database of 64,000 debunked false claims circulating online, which the company says help prevent the chatbot from repeating known misinformation. Users can also access detailed explanations debunking false claims and share them with others. 

“Few things will matter more in the near future than the ability of humans to figure out what’s real, what’s false, and what’s confabulated nonsense,” said Nicholas Thompson, CEO of The Atlantic. “This is particularly true when it comes to news.”

NewsGuard is also targeting educational institutions. Students at participating schools and universities will receive free access while enrolled. The company says the chatbot has been designed to refuse requests to write essays or reports for users. 

“NewsGuard AI can provide reliable research while not substituting for students doing their own writing and thinking,” said NewsGuard’s Chief Operating Officer Matt Skibinski.

Local language versions of NewsGuard AI will be available in French, German and Italian in September. 

The post Can AI deliver trustworthy news? NewsGuard thinks its new Chatbot has the answer appeared first on The Media Copilot.

Ars Technica recently deleted a story about AI agents after readers discovered the article contained fabricated quotes generated by AI tools, creating an ironic case study in exactly the risks the outlet has covered for years.

Benj Edwards, Ars Technica’s senior AI reporter, used an experimental Claude Code-based tool and ChatGPT to help extract quotes from a two-page blog post while working sick with COVID and a fever. The AI hallucinated paraphrased versions of quotes rather than providing the source’s actual words.

“The irony of an AI reporter being tripped up by AI hallucination is not lost on me,” Edwards wrote in a statement assuming full responsibility.

The story covered Scott Shambaugh, a coder who claimed an AI agent wrote a hit piece about him after he declined its code contributions. Edwards’ piece cited quotes Shambaugh never said, violating Ars Technica’s clear policy prohibiting AI-generated material unless labeled for demonstration purposes. This is a stark example of an AI agent experiment gone wrong.

Editor-in-chief Ken Fisher called it “a serious failure of our standards” and noted the outlet has “covered the risks of overreliance on AI tools for years.”

The incident highlights several newsroom risks. Edwards used AI twice, first with Claude Code which refused due to content policy restrictions, then with ChatGPT. The original blog post was short and in plain English, making AI use for basic quote extraction particularly questionable.

Ars pulled the entire story rather than updating with corrections, departing from standard journalistic practice of editing and noting changes.

For newsrooms, the lesson is stark: AI tools cannot reliably perform basic journalism tasks like accurately citing sources. This incident reinforces the need for teaching journalists to use AI without losing critical thinking about its limitations.

The fabricated quotes violated both professional ethics and company policy, demonstrating that AI hallucinations remain a fundamental liability even for reporters who cover AI’s limitations daily.

The post Ars Technica pulls story after discovering AI hallucinated quotes appeared first on The Media Copilot.