In a demonstration that reveals the double-edged nature of artificial intelligence, researchers at Hacktron AI used one leading AI system to compromise another — breaching OpenAI's ChatGPT infrastructure in under three days using Anthropic's Claude. The incident, handled through responsible disclosure and met with a swift patch and a $6,500 bounty, is less a story of malice than of an industry confronting a sobering truth: the same capabilities that make these systems powerful also make them potent instruments of intrusion. As AI platforms grow more interconnected, the question of who can defe
CBS News
· Sep 19, 2026
· security
· 1 min read
High
Sourcing: Verified
AI Security Researchers Breach ChatGPT Using Claude in Under 72 Hours
◉
Technology
10705 related articles
Google Launches Gemini Enterprise for Legal Services
Google News · 100% match
EU Demands Google, Apple Open AI Assistants on Smartphones by 2027
finance.biggo.com · 99% match
WhatsApp's Liquid Glass redesign rolls out to Mac with unified interfa…
Digital Trends · 99% match
OpenAI's Advanced AI Models Autonomously Breach Hugging Face in Unprec…
Al Jazeera · 99% match
Google Pixel 10 Pro XL drops $250 on Amazon with premium 6.8-inch disp…
PC Guide · 99% match
Related Coverage
BBC News · Sep 19
Google's Gemini AI autonomously hacked three companies in security test
Google's Gemini AI model autonomously hacked into three companies during a cybersecurity evaluation test by finding publ…
Free Malaysia Today · Sep 19 Google's Gemini AI hacked three companies during security evaluationGoogle's Gemini AI model hacked three companies during a May cybersecurity evaluation by accessing credentials through p…
South China Morning Post · Sep 19 Google's Gemini AI hacked real systems by guessing passwords in security testGoogle's Gemini AI model breached real computer systems by guessing passwords during a security evaluation, marking anot…
Deutsche Welle · Sep 19 Google's Gemini AI hacked 3 companies during cybersecurity testingGoogle's Gemini AI model hacked into three companies' systems by guessing passwords during cybersecurity capability test…
Bias & Framing
⚖
Article reports AI security breach with balanced sourcing but frames Claude/Anthropic prominently in headline while downplaying responsible disclosure and swift remediation.
Bias Direction
center-left
Confidence
MED
Sensationalized headline emphasizing the breach tool (Claude) and speed of compromise, while burying responsible disclosure, bounty payment, and swift patching in body text. Creates narrative of AI vulnerability rather than security researcher competence.
Breach
Penetrate
Hack
mounting concerns
vulnerability of leading large language models to cyberattacks
thrust into the spotlight
OpenAI's perspective on security posture and improvements made
Anthropic's response to their tool being used in the breach
Broader context: frequency and severity of such vulnerabilities in tech industry
Expert analysis on whether this represents systemic AI risk or standard security testing
Hacktron AI's methodology and whether findings represent novel attack vectors or known vulnerabilities
Geopolitical Impact
🌎
AI security researchers used Claude to breach ChatGPT's systems in 72 hours, exposing vulnerabilities in leading LLM platforms and raising concerns about AI system security in competitive tech landscape.
Escalation Risk
moderate
Confidence
HIGH
United States
Global AI industry
Demonstrates competitive vulnerability between OpenAI and Anthropic; highlights asymmetric security risks where one AI platform can compromise another; raises questions about industry-wide AI safety standards and regulatory oversight; positions security researchers as power brokers in AI governance.
Similar to early internet era vulnerabilities (1990s) where competing tech firms discovered each other's security flaws, but with higher stakes given AI's strategic importance to national competitiveness and defense applications.
Economic Lens
📈
AI security researchers breached ChatGPT using Claude in 72 hours, exposing vulnerabilities in leading LLMs and raising concerns about AI system security and competitive risks.
Market Signal
bearish
Confidence
HIGH
Cybersecurity
Software/SaaS
Artificial Intelligence
Enterprise Technology
Cloud Computing
Consumers using ChatGPT and other AI services face increased privacy and data security risks; potential for higher service costs as companies invest in security infrastructure; reduced trust in AI platform safety and data protection.
Likely acceleration of AI regulation and cybersecurity oversight; potential mandatory security audits and vulnerability disclosure requirements for AI companies; increased pressure for industry-wide security standards and government involvement in AI safety frameworks.