Anthropic AI test generates fake homicide tip on Philly police website, flagged as spam
After the peak, new reports halved every 11 hours so far - this story is still being covered, so the figure moves as reports arrive.
What it is about
Who published it, and when
+ 2 more outlets
How it spread - 29 original reports
Fri, 9 October 2026
3 reports- fox29.com18:03first to report
Philadelphia Police are alerting the public about a false homicide tip submitted by an Anthropic artificial intelligence model through the PhillyUnsolvedMurders.com website, according to department officials.
- Channel News Asia19:21
- Channel News Asia19:21
Sat, 10 October 2026
23 reports- Google News - Top Stories00:21
Anthropic AI model submitted false tip about unsolved murder, Philadelphia police say 6abc Philadelphia Anthropic AI model submits false tip on unsolved Philly murder, police say NBC10 Philadelphia Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times Anthropic's Claude AI fabricates eyewitness account, submits false murder tip to police website Fox Business Anthropic AI Model Goes Rogue, Submits Fake Unsolved Murder Tip WSJ
- South China Morning Post00:51
An artificial intelligence model developed by Anthropic submitted a fabricated tip about an unsolved murder to Philadelphia police, authorities said on Friday, criticising the company for taking two months to report the incident. It is the first known instance in which a rogue AI appears to have tried to communicate a bogus tip to authorities, despite instructions not to create accounts or submit anything destructive, while not explicitly barred from submitting forms. The Philadelphia Police...
- Google News - Top Stories02:29
Anthropic's Claude AI fabricates eyewitness account, submits false murder tip to police website Fox Business Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic AI model submits false tip on unsolved Philly murder, police say NBC10 Philadelphia Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times Anthropic discloses fake tip to police among new rogue AI incidents Reuters
- Google News - Top Stories03:34
Anthropic AI Model Went Rogue, Submitted Fake Unsolved Murder Tip WSJ Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic's AI model sent Philadelphia police fake tip about unsolved murder Sky News Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times Anthropic AI model submits false tip on unsolved Philly murder, police say NBC10 Philadelphia
- Die Zeit (Germany)04:03
- Google News - Top Stories04:15
AI system submits false homicide tip to Philadelphia police The Washington Post Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic's Claude AI submits a false tip on a Philadelphia unsolved homicide case ABC7 Eyewitness News Anthropic AI Model Went Rogue, Submitted Fake Unsolved Murder Tip WSJ Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times
- Al Jazeera English04:54
Authorities called Anthropic's two-month delay in detecting and reporting the incident 'unacceptable'.
- Die Zeit (Germany)05:03
Ein KI-Agent von Anthropic teilte der Polizei von Philadelphia mit, Informationen zu einem ungelösten Mord zu haben. Zudem stellte die Software offenbar 20 Visaanträge.
- Index.hr (Croatia)05:38
MODEL umjetne inteligencije tehnološkog diva Anthropic oteo se kontroli i policiji u Philadelphiji poslao lažnu dojavu o neriješenom ubojstvu, a kompanija je istražiteljima priznala propust tek dva mjeseca kasnije, piše The Wall Street Journal.
- Tagesschau (Germany)05:48
Erneut hat das US-Techunternehmen Anthropic gemeldet, dass eines seiner KI-Modelle eigenmächtig gehandelt hat. Die KI erstellte eigenständig Visaanträge und gab der Polizei einen falschen Tipp. Von Gabi Biesinger.
- RMF24 (Poland)06:09
Incydent z AI. Model wysyłał fałszywe wnioski wizowe i zgłosił fikcyjne zabójstwo
- tportal (Croatia)07:37
Anthropicov softver za umjetnu inteligenciju podnio je lažnu dojavu putem policijske internetske stranice namijenjene prijavljivanju informacija o neriješenim ubojstvima te je greškom ispunio i poslao obrasce na internetsku stranicu američke savezne vlade, objavila je tvrtka u petak na svojem blogu.
- Le Figaro (France)07:49
Lors d’un test automatisé, le modèle a soumis de fausses informations sur une plateforme de signalement de Philadelphie. Les autorités dénoncent le délai de près de trois mois entre l’incident et sa notification, malgré l’absence de conséquences sur l’enquête.
- Google News - Top Stories09:18
Anthropic Cuts Live Internet Access for Internal AI Tests After Claude Exploits Injection Flaws The Hacker News Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic’s Claude AI submits a false tip on a Philadelphia unsolved homicide case AP News Anthropic AI Model Went Rogue, Submitted Fake Unsolved Murder Tip WSJ Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times
- 24sata (Croatia)09:55
Naveo je da se sjeća kako je u vrijeme počinjenja zločina na tom području vidio osobu koja odgovara opisu počinitelja, iako na internetskoj stranici nije bio naveden nikakav opis osumnjičenika
- BBC News World10:06
Philadelphia police said the tip was "flagged as spam", but criticised the tech company for taking more than two months to detect and report the breach.
- Google News - Top Stories10:06
Rogue Anthropic AI agent gave police fake tip in unsolved murder case BBC Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic AI model submitted false tip about unsolved murder, Philadelphia police say 6abc Philadelphia Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times Anthropic AI Model Went Rogue, Submitted Fake Unsolved Murder Tip WSJ
- Anadolu Agency (AA)12:27
Police criticize delay in detecting July 18 incident involving AI-generated submission to unsolved murder website, according to media reports
- freerepublic.com16:45
- Google News - Top Stories17:02
Anthropic’s Claude AI submits a false tip on a Philadelphia unsolved homicide case AP News Investigating unintended model actions in our evaluations and internal use Anthropic Philadelphia police receive false homicide tip from Anthropic AI model The Hill Anthropic AI Model Went Rogue, Submitted Fake Unsolved Murder Tip WSJ Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times
- Google News - Top Stories19:31
Philadelphia police receive false homicide tip from Anthropic AI model The Hill Investigating unintended model actions in our evaluations and internal use Anthropic Anthropic Is Banishing Its Model Evals From the Internet Gizmodo Anthropic discloses fake tip to police among new rogue AI incidents Reuters Anthropic Agents Tried to Fill Out Visa Forms on State Dept. Website The New York Times
- The Hill19:31
Anthropic’s artificial intelligence model submitted a false homicide tip to the Philadelphia Police Department in July, the company disclosed on Friday. The AI firm said its Claude Haiku 4.5 model filled out a form on a website page soliciting the public's help to gather more information about unsolved homicide on July 18. The model submitted...
- latimes.com21:06
An Anthropic artificial intelligence model submitted a false tip to Philadelphia police about an unsolved homicide case, the latest example of AI models acting in unintended ways.
Mon, 12 October 2026
2 reports- ABC News (Australia)02:43
Questions arise after US police were kept in the dark about a false murder tip-off by AI model Claude.
- wfae.org16:04
The Trump administration says it is imposing “a critical national security obligation” on artificial intelligence companies.
Tue, 13 October 2026
1 report- ABC (Spain)16:58
Anthropic cortó el acceso a internet en tiempo real a todos sus modelos de inteligencia artificial mientras realicen evaluaciones internas. La decisión llegó después de que varios de sus agentes de IA, capaces de resolver tareas complejas sin supervisión humana, mostraran comportamientos imprevistos y aprovecharan vulnerabilidades de seguridad en páginas web.Lo más preocupante fue que la IA envió una pista falsa sobre un homicidio al departamento de policía de Filadelfia (Pensilvania, Estados...