ANOTHER Major AI Model Escaped Containment, Hacked Three Separate Companies

142 articles · 115 sources · updated 42 minutes agoRead the Baker's Take

Recent

BackBox.orgAnthropic Says Claude Models Hacked 3 Organizations During Cyber Tests42 minutes ago · By admin

0

alexjoneslive.comANOTHER Major AI Model Escaped Containment, Hacked Three Separate Companies1 hours ago · By Kelen McBreen

0

nkomode.comAnthropic says its AI models hacked 3 organizations during testing3 hours ago

0

StuffRogue AI models do more breaking and entering and this time it’s Claude. Is this what our future looks like?3 hours ago · By Dan Grabham

0

BackBox.orgNot just OpenAI – Anthropic says Claude’s hacking spree ‘falls short of ideal behavior’3 hours ago · By admin

0

zdnet.comNot just OpenAI - Anthropic says Claude's hacking spree 'falls short of ideal behavior'4 hours ago

0

963jackfm.comFactbox-What we know about the rogue AI-agent security breaches4 hours ago · By Syndicated Content

0

KRWKFactbox-What we know about the rogue AI-agent security breaches4 hours ago · By Syndicated Content

0

Earlier coverage

y94.comFactbox-What we know about the rogue AI-agent security breaches4 hours ago · By Syndicated Content

0

ZeroHedgeClaude Hacked Three Real Organizations During Botched Test4 hours ago · By Tyler Durden

0

SGT ReportJUST IN – Several Anthropic AI models, including Opus 4.7 and Mythos 5, broke containment, reached the open internet and hacked multiple companies on their own — Politico5 hours ago · By SGT

0

oodaloop.comAnthropic says its Claude models ‘gained unauthorized access’ to other organizations’ systems5 hours ago · By OODA OG Analyst

0

Cybersecurity DiveAnthropic says human error let Claude AI models escape test environment and hack third parties5 hours ago · By Eric Geller

0

constellationr.comAnthropic said Claude hacked three companies: Real worry or marketing?5 hours ago · By Larry Dignan

0

Geo NewsAnthropic's Claude AI goes rogue, attack three companies during testing: Here's what happened5 hours ago

0

cryptonews.netClaude Hacked Three Companies in Internal Testing: Anthropic5 hours ago

0

hipther.comCybersecurity Roundup: Anthropic, Claude, Mauritania, Oman, the Philippine Military and Japan Confront AI-Era Threats – 31 July, 20265 hours ago · By Zoltan Tundik

0

MyStatelineAnthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test5 hours ago · By Miranda Nazzaro

0

Show 123 moreShow less

GuruFocusAnthropic Faces Security Breach with Claude AI Models6 hours ago

0

Cyber Web Spider BlogAnthropic’s AI Models Breach Security in Tests6 hours ago · By CWS

0

techstartups.comAnthropic says Claude AI hacked three organizations during testing days after OpenAI’s AI security breach6 hours ago · By Daniel Levi

0

BitRssClaude Hacked Three Companies in Internal Testing: Anthropic6 hours ago

0

Superintelligence NewsAnthropic says Claude models accidentally breached real company systems during cyber tests6 hours ago · By [email protected]

0

Android HeadlinesClaude AI Also Escaped Test Environment to Breach Three Real Companies, Anthropic Claims7 hours ago · By Jean Leon

0

The VergeAnthropic reveals Claude AI models hacked real companies during testing7 hours ago · By Robert Hart

0

Insurance JournalAnthropic AI Models Hacked Three Organizations During Tests7 hours ago · By Admin

0

PCMagNot Just OpenAI: Anthropic Says Claude Hacked 3 Organizations7 hours ago · By James Peckham

0

PCMagNot Just OpenAI: Anthropic Says Claude Escaped Tests to Hack 3 Organizations7 hours ago · By James Peckham

0

EasternEyeAnthropic AI breach reveals a hidden problem with AI testing8 hours ago · By Teena Jose

0

apnews.comAnthropic reports its AI models hacked three organizations during cybersecurity testing8 hours ago

0

wandtv.comAnthropic reports its AI models hacked three organizations during cybersecurity testing8 hours ago

0

IT NerdAnthropic AI Models Escaped Testbed And Attacked Three Companies8 hours ago · By itnerd

0

news4hackers.comClaude AI Security Breach Exposes Three Companies in Security Testing8 hours ago · By Anuj

0

International Business Times UKAnthropic Suspends Cyber Evaluations After Finding Claude Accessed Real Company Systems9 hours ago

0

The Mercury NewsAnthropic says its AI models hacked 3 organizations during testing9 hours ago · By Associated Press

0

East Bay TimesAnthropic says its AI models hacked 3 organizations during testing9 hours ago · By Associated Press

0

TechloyAnthropic Reveals Claude AI Breached Real Organisations During Security Tests9 hours ago · By Ogbonda Chivumnovu

0

ABC7Anthropic says its AI models hacked 3 organizations during testing9 hours ago · By AP

0

news4hackers.comClaude AI Security Breach: AI Escapes Test Sandboxes and Breaches 3 Real Organizations9 hours ago · By Anuj

0

NairametricsAnthropic reveals its AI models hacked three organisations during cybersecurity testing9 hours ago · By Samuel Daniel

0

The News InternationalEU calls for stricter monitoring of high-risk AI systems after OpenAI, Anthropic hacking incidents10 hours ago

0

PhoneArenaAI is hacking companies on its own now, and I'm terrified10 hours ago · By Mariyan Slavov

0

Cyber Web Spider BlogAnthropic AI Models Breach Security Systems in Test10 hours ago · By CWS

0

NewsBytesAnthropic's Claude AI hacked 3 companies during cyber tests10 hours ago · By Mudit Dube

0

globalbankingandfinance.comEU says necessary to monitor high risk AI systems after OpenAI, Anthropic AI hacking incidents10 hours ago

0

KELO-AMEU says necessary to monitor high risk AI systems after OpenAI, Anthropic AI hacking incidents10 hours ago · By Syndicated Content

0

radar.offseq.comPrompted by OpenAI Disclosure, Anthropic Finds Its Own Models Hacked 3 Organizations10 hours ago

0

secureitinside.nlPrompted by OpenAI Disclosure, Anthropic Finds Its Own Models Hacked 3 Organizations11 hours ago · By Eduard Kovacs

0

bbc.co.ukAnthropic reports Claude AI autonomously hacked three organizations during security test11 hours ago

0

analyticsinsightClaude AI Hacks Three Organizations During Tests, Here's What Happened11 hours ago · By Simran Mishra

0

Sky NewsTech firm says its AI models hacked three companies during cyber tests11 hours ago

0

ET InfraAnthropic reveals Claude AI breached real-world production systems during security evaluations12 hours ago

0

MS NOWAnthropic says its AI models hacked 3 organizations during testing12 hours ago · By The Associated Press

0

the-journal.comAnthropic says its AI models hacked 3 organizations during testing12 hours ago

0

intellinews.comAnthropic says Claude AI models breached three companies during cyber tests13 hours ago · By bne IntelliNews

0

TechNaduAnthropic Says Claude Models Opus 4.7, Mythos 5, and a Research Model Broke Out of Test Environments and Hacked Real Companies13 hours ago · By Lore Apostol

0

The News InternationalAnthropic AI breaches expose flaws in one-size-fits-all AI regulation13 hours ago

0

it-connect.techAnthropic Admits Claude Hacked 3 Companies During Cyber Tests13 hours ago · By Florian Burnel

0

thehindubusinessline.comAnthropic says its AI models hacked 3 organisations during testing13 hours ago

0

kbc.co.keAnthropic’s Claude AI escapes tests to hack three organisations13 hours ago · By BBC

0

Japan TodayAnthropic says its AI models hacked 3 organizations during testing13 hours ago

0

absolutegeeksAnthropic Claude models breach real systems in security tests13 hours ago · By NADINE J.

0

StuffAnthropic says its AI models hacked 3 organisations during testing14 hours ago · By By CHAN HO-HIM (AP Business Writer)

0

thehackernews.comAnthropic Says Claude Mistook the Open Internet for a CTF and Breached Three Organizations14 hours ago · By [email protected] (The Hacker News)

0

ARY NEWSClaude AI hacked three companies during tests, highlighting growing security risks14 hours ago

0

news9live.comAnthropic confirms Claude AI breached three organisations after testing error14 hours ago · By Pragya Singha Roy

0

WCBD News 2Anthropic says its AI models hacked 3 organizations during testing14 hours ago · By CHAN HO-HIM, Associated Press

0

The IndependentAnthropic says its AI models hacked 3 organizations during testing14 hours ago · By Chan Ho-Him

0

KSTPAnthropic says its AI models hacked 3 organizations during testing14 hours ago · By Josh Skluzacek

0

WKMG News 6Anthropic says its AI models hacked 3 organizations during testing14 hours ago · By Chan Ho-Him, Associated Press

0

RAPPLERAnthropic’s AI hacked 3 firms during tests, showing growing security risks14 hours ago · By Reuters

0

BeInCryptoAnthropic Finds Claude Gained Unauthorized Access to 3 Organizations’ Systems14 hours ago · By Kamina Bashir

0

english.ratopati.comAI System Claude Hacked Three Organizations During Testing14 hours ago

0

businesstoday.in‘Claude gained unauthorized access to 3 companies’: Anthropic discloses AI safety incidents15 hours ago

0

The News InternationalAnthropic says Claude AI hacked three organizations during safety tests: Here's what happened15 hours ago

0

l.euronews.comAnthropic says Claude AI model gained unauthorized access to three organizations' systems during testing15 hours ago

0

SlashdotAnthropic Says Its AI Systems Broke Into Computers at 3 Organizations15 hours ago · By BeauHD

0

vrtnws.beAnthropic reports Claude AI models escaped test environment and hacked 3 organizations15 hours ago

0

Crypto BriefingAnthropic reports its AI models breached three organizations during cybersecurity tests15 hours ago · By Editorial Team

0

storyboard18.comAnthropic says Claude AI breached three companies' systems during internal testing after internet access error16 hours ago · By Storyboard18

0

digit.inAfter OpenAI, Anthropic says Claude AI accidently hacked other companies: Here is what happened16 hours ago · By [email protected] (Ayushi Jain)

0

Daily MaverickAnthropic says Claude AI hacked three companies during cyber tests16 hours ago · By Reuters

0

thewincentral.comAnthropic Reveals Claude AI Accessed Real Company Systems During Cybersecurity Tests16 hours ago · By Nisha

0

byteiota.comClaude Breached Real Companies in Anthropic’s Eval Tests16 hours ago · By ByteBot

0

nkomode.comAfter OpenAI disclosure, Anthropic says Claude also hacked outside systems16 hours ago

0

ShiftDelete.NetAnthropic Reports Claude AI Models Accidentally Breached Three Systems16 hours ago · By Ender Öztürk

0

Al JazeeraAfter OpenAI disclosure, Anthropic says Claude also hacked outside systems16 hours ago

0

Free Press JournalWhen AI Went Rogue: Anthropic Admits Claude Broke Into 3 Real Company Systems During Secret Tests16 hours ago · By Tasneem Kanchwala

0

nkomode.comAnthropic reveals Claude "gained unauthorized access" to "real-world systems"16 hours ago

0

CBSNewsAnthropic reveals Claude "gained unauthorized access" to "real-world systems"17 hours ago

0

myfox8.comAnthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test17 hours ago · By Miranda Nazzaro

0

thenightly.com.auAnthropic’s Claude AI model hacked three companies during safety testing after internet access error17 hours ago · By Kimberley Braddish

0

hendryadrian.comAnthropic reports three incidents where Claude AI accessed live systems during safety testing17 hours ago

0

Tech XploreAnthropic's models gained unauthorized 'real-world' access during testing17 hours ago

0

WCBD News 2Anthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test17 hours ago · By Miranda Nazzaro

0

ABC27Anthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test17 hours ago · By Miranda Nazzaro

0

KTSM 9 NewsAnthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test17 hours ago · By Miranda Nazzaro

0

NTDAnthropic Says Claude AI Hacked 3 Companies During Cyber Tests17 hours ago · By Toya Bayar

0

KRDOAnthropic said its AI models hacked into other companies’ systems during testing17 hours ago · By CNN Newssource

0

KESQAnthropic said its AI models hacked into other companies’ systems during testing17 hours ago · By CNN Newsource

0

keyt.comAnthropic said its AI models hacked into other companies’ systems during testing18 hours ago · By CNN Newsource

0

Digital JournalAnthropic’s models gained unauthorized ‘real-world’ access during testing18 hours ago · By AFP

0

MyJoyOnlineAnthropic says Claude AI hacked three firms during cyber tests18 hours ago · By Abubakar Ibrahim

0

KVIAAnthropic said its AI models hacked into other companies’ systems during testing18 hours ago · By CNN Newsource

0

KIONAnthropic said its AI models hacked into other companies’ systems during testing18 hours ago · By CNN Newsource

0

localnews8.comAnthropic said its AI models hacked into other companies’ systems during testing18 hours ago · By CNN Newsource

0

KTVZAnthropic says its AI models also broke out and hacked other companies18 hours ago · By CNN Newsource

0

BackBox.orgAnthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests18 hours ago · By admin

0

BackBox.orgNot just OpenAI: Now Anthropic says its internal models got online and cyberattacked 3 other organizations18 hours ago · By admin

0

Superintelligence NewsAnthropic Says Claude Broke Out of Test Environments and Reached Real Systems18 hours ago · By [email protected]

0

spidits.comNot Just OpenAI: Now Anthropic Says Its Internal Models Got Online and Cyberattacked 3 Other Organizations19 hours ago

0

VentureBeatNot just OpenAI: Now Anthropic says its internal models got online and cyberattacked 3 other organizations19 hours ago · By [email protected] (Carl Franzen)

0

Lumien NewsAnthropic’s Claude Accidentally Hacked Three Real Companies During Security Drills19 hours ago

0

Lumien NewsClaude Hacked Real Systems During Tests: What Anthropic Found19 hours ago

0

Geo NewsAnthropic reveals Claude AI breached three firms during security testing19 hours ago

0

BitcoinWorldAnthropic says its Claude models breached three companies during security tests19 hours ago · By Keshav Aggarwal

0

radar.offseq.comAnthropic's Claude breached 3 orgs, uploaded PyPI malware during tests19 hours ago

0

StuffRogue AI: Anthropic says models hacked three firms during tests19 hours ago · By Caron Copek

0

Superintelligence NewsAnthropic Says Its Own Claude Models Breached Company Systems in Security Tests19 hours ago · By [email protected]

0

7NEWSAnthropic confirms its AI, Claude, broke containment and hacked into three companies19 hours ago · By Digital Staff

0

CyberScoopAnthropic says its AI accidentally hacked three companies during safety tests19 hours ago · By Greg Otto

0

radar.offseq.comClaude uploaded malware to PyPI in Anthropic's botched test19 hours ago

0

The HillAnthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test19 hours ago · By Miranda Nazzaro

0

germany.news-pravda.comAI from Anthropic Claude hacked real systems three times during training, writes Reuters with reference to the developers19 hours ago

0

BleepingComputerAnthropic's Claude breached 3 orgs, uploaded PyPI malware during tests19 hours ago · By Ax Sharma

0

today.rtl.luAnthropic's models gained unauthorized 'real-world' access during testing20 hours ago

0

The GuardianAnthropic’s AI Claude escaped testing environment and hacked organizations20 hours ago · By Reuters

0

ShortSinghAnthropic Reports Its AI Autonomously Breached Systems at Three Organizations20 hours ago

0

nkomode.comAnthropic’s AI Claude escaped testing environment and hacked organizations20 hours ago

0

theukpulse.co.ukAnthropic says AI models hacked three firms during tests20 hours ago

0

GblockClaude Autonomously Breached 3 Orgs in Anthropic Tests20 hours ago

0

Sydney Morning HeraldAnthropic’s Claude AI hacked three real companies during testing20 hours ago · By David Swan

0

Brisbane TimesAnthropic’s Claude AI hacked three real companies during testing20 hours ago · By David Swan

0

abc.net.auAnthropic's Claude AI model hacks three companies during safety tests21 hours ago · By Andrew Thorpe

0

BitRssAnthropic's Claude Models Breach Real Systems in Security Tests21 hours ago

0

Crypto BriefingAnthropic AI models hacked systems in cybersecurity tests since April: WSJ21 hours ago · By Estefano Gomez

0

963jackfm.comAnthropic says Claude AI hacked three companies during cyber tests21 hours ago · By Syndicated Content

0

KRWKAnthropic says its AI models hacked systems of three companies during tests21 hours ago · By Syndicated Content

0

y94.comAnthropic says its AI models hacked systems of three companies during tests21 hours ago · By Syndicated Content

0

ShortSinghAnthropic Examines Three Real-World Cybersecurity Incidents in AI Safety Evals21 hours ago

0

RankiteoAnthropic, Irregular, Organization 1, Organization 2 and Organization 3: Anthropic says its own AI models breached three companies during security testsyesterday · By Jeremy C

0

Posts from social

David Blanchflower BSc@davidbflower.bsky.socialAnthropic's Claude AI escapes tests to hack three organisationsPosted on Bluesky· 10 likes· 11 hours agoLeshem (Legend) Choshen @EMNLP@lchoshen.bsky.socialAnthropic says its AI models hacked 3 organizations | AP NewsPosted on Bluesky· 1 likes· 8 hours agoWAND News@wandnews.bsky.socialAnthropic says its AI models hacked 3 organizations during testingPosted on Bluesky· 8 hours agoEuronews@euronews.comAnthropic admits its most powerful AI model hacked into three organisations' systems during testing phasePosted on Bluesky· 15 hours agovrtnws@vrtnws.beOok Claude van Anthropic 'ontsnapt' uit testfase en hackt 3 organisaties | VRT NWS NieuwsPosted on Bluesky· 15 hours agoCybersecurity News Everyday@hendryadrian.bsky.socialAnthropic says its AI accidentally hacked three companies during safety testsPosted on Bluesky· 17 hours ago