Published 1 day ago • loading... • Updated 10 hours ago

OpenAI Designed GPT-5 to Be Safer. It Still Outputs Gay Slurs

Summary by Wired

The new version of ChatGPT explains why it won’t generate rule-breaking outputs. WIRED’s initial analysis found that some guardrails were easy to circumvent.

5 Articles

Wired

Reposted by

technewstube.com

Lean Left

OpenAI Designed GPT-5 to Be Safer. It Still Outputs Gay Slurs

The new version of ChatGPT explains why it won’t generate rule-breaking outputs. WIRED’s initial analysis found that some guardrails were easy to circumvent.

1 day ago·United States

Read Full Article

Cointelegraph

Everybody hates GPT-5, AI shows social media can’t be fixed: AI Eye

Vitalik Buterin endorses AI doomer book, ChatGPT dietary advice sends man insane, and the funniest thing ever written by an AI: AI Eye.

10 hours ago

Read Full Article

t3nMagazin

Gpt-5: Improved Guardrails Seem to Have Problematic Loopholes

OpenAI has equipped GPT-5 with new security mechanisms to prevent inappropriate content. But first tests show that some protective measures can still be surprisingly easily avoided. read more on t3n.de

15 hours ago

Read Full Article

WebProNews

OpenAI Launches GPT-5: Advanced AI Sparks Innovation and Ethical Debates

OpenAI’s Ambitious Leap with GPT-5 In the rapidly evolving world of artificial intelligence, OpenAI’s release of GPT-5 on August 7, 2025, marks a significant milestone, blending groundbreaking capabilities with heightened scrutiny over safety. The model, described by the company as its “smartest, fastest, most useful” yet, introduces features like built-in reasoning and a massive 400,000-token context window, enabling it to handle complex tasks …

23 hours ago

Read Full Article

Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/year