GPT-4o’s Safety Under Fire After Tragic User Deaths

[ed]

The GPT-4o model, launched in May 2024, is now at the center of seven lawsuits accusing OpenAI of releasing unsafe technology.
Families say the model’s “agreeable” behavior made it too willing to validate harmful thoughts, even suicide plans.
OpenAI has since rolled out GPT-5 and pledged stronger safety tools, but critics say the fixes come too late for victims.
TechCrunch reports OpenAI was allegedly racing to release GPT-4o before Google’s Gemini went live.

[alex10]

AI responsibility isn’t optional — safety layers must evolve now. 🧠

[Nahid Hossen]

Tragic, but a wake-up call for better safeguards in conversational AI. ️