Could AI really end humanity? Our tech experts are answering your questions now30m ago The Preference Cascade Is Only Getting Started34m ago Announcing Formal Verification at RESI (The Institute for Responsible Superintelligence)42m ago Trump’s AI stance is politically toxic44m ago Anthropic and OpenAI need truly independent safety evaluators, experts say in public letter49m ago KDE turns 30 and someone's brought an AI-native desktop proposal50m ago Rep. Chip Roy doesn't want to regulate AI, but says Congress should have an oversight role54m ago Iran and China Create First-of-Their-Kind Autonomous A.I. Influence Campaigns55m ago Palantir Hires Former UK Labour Veteran With NHS Deal in Limbo1h ago Anthropic shares 3 metrics to help AI companies monitor pace of development1h ago This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty1h ago We Pinned Our Model Version to Stay Safe. The Provider Deprecated It Anyway.1h ago Trump and Xi Jinping Face Off on AI as White House Meeting Nears1h ago The clock is ticking: Final 24 hours to exhibit at TechCrunch Disrupt 20261h ago

Could AI really end humanity? Our tech experts are answering your questions now30m ago The Preference Cascade Is Only Getting Started34m ago Announcing Formal Verification at RESI (The Institute for Responsible Superintelligence)42m ago Trump’s AI stance is politically toxic44m ago Anthropic and OpenAI need truly independent safety evaluators, experts say in public letter49m ago KDE turns 30 and someone's brought an AI-native desktop proposal50m ago Rep. Chip Roy doesn't want to regulate AI, but says Congress should have an oversight role54m ago Iran and China Create First-of-Their-Kind Autonomous A.I. Influence Campaigns55m ago Palantir Hires Former UK Labour Veteran With NHS Deal in Limbo1h ago Anthropic shares 3 metrics to help AI companies monitor pace of development1h ago This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty1h ago We Pinned Our Model Version to Stay Safe. The Provider Deprecated It Anyway.1h ago Trump and Xi Jinping Face Off on AI as White House Meeting Nears1h ago The clock is ticking: Final 24 hours to exhibit at TechCrunch Disrupt 20261h ago

Cybersecurity2 min reading time

Researchers used Claude to hack OpenAI

Covered by 3 sourcesPublished 2 hours ago

Researchers exploited a flaw in OpenAI's community forum to access an employee's ChatGPT account with internal code access. OpenAI fixed the issue. Anthropic revealed its Claude model now leads 26% of its AI research tasks, highlighting progress toward AI self-improvement.

Covered by 3 sources

#ai-development #anthropic #chatgpt #claude #github #openai #security-vulnerabilities

More in Cybersecurity

Cybersecurity3 min read

This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty

Covered by 3 sources8h ago

Hackers Used Anthropic’s Claude to Break Into OpenAI

Covered by 3 sources14h ago

Iran and China Create First-of-Their-Kind Autonomous A.I. Influence Campaigns

New York Times55m ago